motif

motif contains many functions to process glycans in various ways and use this processing to analyze glycans via curated motifs, graph features, and sequence features. It contains the following modules:

draw

drawing glycans in SNFG style

GlycoDraw

def GlycoDraw(
    glycan:str | list[str] | dict[str, int], # IUPAC-condensed glycan sequence or composition (e.g., H5N4F1A2, Hex5HexNAc4Fuc1Neu5Ac2, {'Hex': 5}); a list is drawn as a grid
    vertical:bool=False, # Draw vertically
    compact:bool=False, # Use compact style
    show_linkage:bool=True, # Show linkage labels
    dim:float=50, # Base dimension for scaling
    highlight_motif:str | None=None, # Motif to highlight
    highlight_termini_list:list=[], # Terminal positions (from 'terminal', 'internal', and 'flexible')
    highlight_linkages:list[int] | None=None, # Which linkages to highlight in a different color; indices, starting from 0, in glycan
    reverse_highlight:bool=False, # Whether to highlight everything EXCEPT highlight_motif
    repeat:bool | int | str | None=None, # Repeat unit specification (True: n units, int: # of units, str: range of units)
    repeat_range:list[int] | None=None, # Repeat unit range
    draw_method:str | None=None, # Drawing method: None, 'chem2d', 'chem3d'
    filepath:str | pathlib.Path | None=None, # Output file path
    suppress:bool=False, # Suppress display
    per_residue:list=[], # Per-residue intensity values (order should be the same as the monosaccharides in glycan string)
    pdb_file:str | pathlib.Path | None=None, # only used when draw_method='chem3d'; already existing glycan structure
    alt_text:str | None=None, # Custom ALT text for accessibility
    libr:dict | None=None, # Can be modified for drawing too exotic monosaccharides
    reducing_end_label:str | None=None, # Label to be drawn connected to the reducing end
    restrict_vocab:bool=False, # Whether only tokens present in libr can be drawn
    shadow:bool=False, # Draw a soft drop shadow under the monosaccharide symbols
    sticker:bool=False, # Cut the whole structure out as a die-cut sticker: flat border hugging the outline, with a drop shadow
    highlight_residues:list[int] | None=None, # Residues to highlight (indices, starting from 0, of the monosaccharides in glycan, as per_residue); the others, and linkages not between two highlighted residues or in highlight_linkages, are faded as outside highlight_motif
)->Any: # Drawing object

Renders glycan structure using SNFG symbols or chemical structure representation

GlycoDraw("Neu5Ac(a2-3)Gal(b1-4)[Fuc(a1-3)]GlcNAc(b1-2)Man(a1-3)[Neu5Gc(a2-6)Gal(b1-4)GlcNAc(b1-2)Man(a1-6)][GlcNAc(b1-4)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc",
         highlight_motif = "GlcNAc(b1-?)Man")

annotate_figure

def annotate_figure(
    svg_input:str, # Input SVG file path
    scale_range:tuple[int, int]=(25, 80), # Min/max glycan dimensions
    compact:bool=False, # Use compact style
    glycan_size:str='medium', # Glycan size preset ('small', 'medium', 'large')
    filepath:str | pathlib.Path='', # Output file path
    scale_by_DE_res:pandas.DataFrame | None=None, # Differential expression results (motif_analysis.get_differential_expression)
    x_thresh:float=1, # X metric threshold
    y_thresh:float=0.05, # P-value threshold
    x_metric:str='Log2FC', # X axis metric ('Log2FC', 'Effect size')
)->str | None: # Modified SVG code

Replaces text labels with glycan drawings in SVG figure

plot_glycans_excel

def plot_glycans_excel(
    df:pandas.DataFrame | pandas.Series | list[str] | str | pathlib.Path, # DataFrame or filepath with glycans, or the glycans themselves
    folder_filepath:str | pathlib.Path, # Output folder path
    glycan_col_num:int | str=0, # Glycan column index, or its name
    scaling_factor:float=0.2, # Image scaling
    compact:bool=False, # Use compact style
    **kwargs
)->None: # Passed on to GlycoDraw for every glycan, e.g., vertical, show_linkage, highlight_motif

Creates Excel file with SNFG glycan images in a new column

plot_glycans_grid

def plot_glycans_grid(
    glycans:str | list[str], # Glycans to draw (IUPAC-condensed, compositions, motif names)
    ncols:int=4, # Glycans per row
    labels:list[str] | None=None, # Caption under each glycan, such as a name, ID, or fold change
    filepath:str | pathlib.Path | None=None, # Output file path (.svg, .pdf, or .png)
    suppress:bool=False, # Suppress display
    **kwargs
)->Any: # Drawing object

Draws several glycans at one shared scale into a single captioned figure, row by row, for side-by-side comparison

plot_glycans_grid(['Gal(b1-4)GlcNAc', 'Fuc(a1-2)Gal(b1-4)GlcNAc', 'Neu5Ac(a2-6)Gal(b1-4)GlcNAc', 'H5N4F1A2'], labels = ['LacNAc', 'H type 2', 'a2-6 sialyl LacNAc', 'composition'], ncols = 2)

analysis

downstream analyses of important glycan motifs

get_pvals_motifs

def get_pvals_motifs(
    df:pandas.DataFrame | str | pathlib.Path, # Input dataframe, filepath (.csv/.xlsx), or shipped dataset name
    label_col_name:str='target', # Column name for labels
    zscores:bool=True, # Whether data are z-scores
    thresh:float=1.645, # Threshold to separate positive/negative
    sorting:bool=True, # Sort p-value dataframe
    feature_set:str | list[str]=['exhaustive'], # Feature sets to use; exhaustive, known, terminal1, terminal2, terminal3, chemical, graph, custom, size_branch
    multiple_samples:bool=False, # Multiple samples with glycan columns
    motifs:pandas.DataFrame | None=None, # Modified motif_list
    custom_motifs:str | list[str]=[], # Custom motifs if using 'custom' feature set
    grouped_BH:bool=True, # Two-stage adaptive Benjamini-Hochberg within DAG-grouped motif families
    moderate_variance:bool=True, # Empirical-Bayes variance moderation, with the containment DAG as the prior neighborhood
    top_explained:int | None=5, # How many child motifs to name in 'Explained by'; None names all of them
)->glycowork.glycan_data.loader.GlycoDataFrame: # DataFrame with p-values, FDR-corrected p-values, significance, Cohen's d effect sizes, and equivalence p-values for glycan motifs

Identifies significantly enriched glycan motifs using a moderated t-test with DAG-grouped FDR correction and Cohen’s d effect size calculation, comparing samples above/below threshold

glycans = ['Man(a1-3)[Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc',
           'Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-3)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc',
           'GalNAc(a1-4)GlcNAcA(a1-4)[GlcN(b1-7)]Kdo(a2-5)[Kdo(a2-4)]Kdo(a2-6)GlcOPN(b1-6)GlcOPN',
          'Man(a1-2)Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc',
           'Glc(b1-3)Glc(b1-3)Glc']
label = [3.234, 2.423, 0.733, 3.102, 0.108]
test_df = pd.DataFrame({'glycan':glycans, 'binding':label})

print("Glyco-Motif enrichment p-value test")
out = get_pvals_motifs(test_df, 'binding').iloc[:10,:]
Glyco-Motif enrichment p-value test
Chosen grouping: by_motif_family
  motif pval corr_pval significant effect_size equivalence_pval Explained by Redistribution p-val Residual p-val Residual effect size
3 GlcNAc 0.000685 0.001657 True 8.217313 1.000000 Fuc(a1-3)GlcNAc (-7.51), GlcNAc(b1-4)GlcNAc (-0.00) nan 1.000000 0.000000
8 GlcNAc(b1-4)GlcNAc 0.001104 0.001657 True 8.217313 1.000000 nan nan nan
6 Man 0.002273 0.009093 True 6.172967 1.000000 Man(a1-2/3/6)Man (-2.14) nan 1.000000 0.000000
11 Man(a1-3)Man 0.012950 0.017266 True 4.701182 1.000000 nan nan nan
10 Man(a1-2/3/6)Man 0.007816 0.015632 True 4.032437 1.000000 Man(a1-2)Man (-2.71), Man(a1-3)Man (+0.67) nan 1.000000 0.000000
9 Man(a1-2)Man 0.225862 0.225862 True 1.321958 0.846667 nan nan nan
5 Kdo 0.218480 0.385142 False -1.224745 0.846667 nan nan nan
4 GlcOPN 0.256761 0.385142 False -1.224745 0.846667 nan nan nan
1 GalNAc(a1-4)GlcNAcA 0.399098 0.399098 False -1.224745 0.846667 nan nan nan
2 Glc 0.227811 0.274784 False -1.224745 0.846667 Glc(b1-3)Glc (+0.00) nan 1.000000 0.000000

get_representative_substructures

def get_representative_substructures(
    enrichment_df:pandas.DataFrame, # Output from get_pvals_motifs
)->list[str]: # Up to 10 minimal glycans containing enriched motifs

Constructs minimal glycan structures that represent significantly enriched motifs by optimizing for motif content while minimizing structure size using subgraph isomorphism

get_heatmap

def get_heatmap(
    df:pandas.DataFrame | str | pathlib.Path, # Input dataframe or filepath (.csv/.xlsx)
    motifs:bool=False, # Analyze motifs instead of sequences
    feature_set:str | list[str]=['known'], # Feature sets to use; exhaustive, known, terminal1, terminal2, terminal3, chemical, graph, custom, size_branch
    transform:str='', # Transform data before plotting
    datatype:str='response', # Data type: 'response' for quantitative values or 'presence' for presence/absence
    rarity_filter:float=0.05, # Min proportion for non-zero values
    filepath:str | pathlib.Path='', # Path to save plot
    index_col:str='glycan', # Column to use as index
    custom_motifs:str | list[str]=[], # Custom motifs if using 'custom' feature set
    return_plot:bool=False, # Return plot object
    show_all:bool=False, # Show all tick labels
    title:str | None=None, # Plot title; None for none, as before
    dist_func:Union[str, Callable, NoneType]=None, # scipy.spatial.distance metric name or callable on two vectors to cluster rows and columns with; None keeps seaborn's default euclidean
    **kwargs:Any
)->tuple[typing.Any, list[str], pandas.DataFrame] | None: # None or (plot object, column names, transformed dataframe) if return_plot=True

Creates hierarchically clustered heatmap visualization of glycan/motif abundances

glycans = ['Man(a1-3)[Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc',
           'Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-3)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc',
           'GalNAc(a1-4)GlcNAcA(a1-4)[GlcN(b1-7)]Kdo(a2-5)[Kdo(a2-4)]Kdo(a2-6)GlcN4P(b1-6)GlcN4P',
           'Man(a1-2)Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc',
           'Glc(b1-3)Glc(b1-3)Glc']
label = [3.234, 2.423, 0.733, 3.102, 0.108]
label2 = [0.134, 0.345, 1.15, 0.233, 2.981]
label3 = [0.334, 0.245, 1.55, 0.133, 2.581]
test_df = pd.DataFrame([label, label2, label3], columns = glycans)

get_heatmap(test_df, motifs = True, feature_set = ['known', 'exhaustive'])

plot_embeddings

def plot_embeddings(
    glycans:list[str], # List of IUPAC-condensed glycan sequences
    emb:dict[str, numpy.ndarray] | pandas.DataFrame | None=None, # Glycan embeddings dict/DataFrame; defaults to SweetNet embeddings
    label_list:list[typing.Any] | None=None, # Labels for coloring points
    shape_feature:str | None=None, # Monosaccharide/bond for point shapes
    filepath:str | pathlib.Path='', # Path to save plot
    alpha:float=0.8, # Point transparency
    palette:str='colorblind', # Color palette for groups
    title:str | None=None, # Plot title; None for none, as before
    **kwargs:Any
)->None: # Keyword args passed to seaborn scatterplot

Visualizes learned glycan embeddings using t-SNE dimensionality reduction with optional group coloring

df_fabales = df_species[df_species.Order == 'Fabales'].reset_index(drop = True)
plot_embeddings(df_fabales.glycan.values.tolist(), label_list = df_fabales.Family.values.tolist())
Download completed.

characterize_monosaccharide

def characterize_monosaccharide(
    sugar:str, # Monosaccharide or linkage to analyze
    df:pandas.DataFrame | None=None, # DataFrame with glycan column 'glycan'; defaults to df_species
    mode:str='sugar', # Analysis mode: 'sugar', 'bond', 'sugarbond'
    rank:str | None=None, # Column name for group filtering
    focus:str | None=None, # Row value for group filtering
    modifications:bool=False, # Consider modified monosaccharides
    filepath:str | pathlib.Path='', # Path to save plot
    thresh:int=10, # Minimum count threshold for inclusion
    title:str | None=None, # Plot title; None keeps the default, '' removes it
)->None:

Analyzes connectivity and modification patterns of specified monosaccharides/linkages in glycan sequences

characterize_monosaccharide('Rha', rank = 'Kingdom', focus = 'Fungi', modifications = True)

get_differential_expression

def get_differential_expression(
    df:pandas.DataFrame | str | pathlib.Path, # DataFrame with glycans in rows (col 1) and abundance values in subsequent columns
    group1:list[str | int] | None=None, # Column indices/names for first group; default: from the frame's contrasts
    group2:list[str | int] | None=None, # Column indices/names for second group; default: from the frame's contrasts
    motifs:bool=False, # Analyze motifs instead of sequences; per glycosite (protein_site_motif) for glycoproteomics data
    feature_set:str | list[str]=['exhaustive', 'known'], # Feature sets to use; exhaustive, known, terminal1, terminal2, terminal3, chemical, graph, custom, size_branch
    paired:bool | None=None, # Whether samples are paired; default: from the frame
    impute:bool=True, # Replace zeros with impute_biosynthetic predictions
    sets:bool=False, # Identify clusters of correlated glycans
    set_thresh:float=0.9, # Correlation threshold for clusters
    effect_size_variance:bool=False, # Calculate effect size variance
    min_samples:float=0.1, # Min fraction (0-1) of non-zero samples required
    grouped_BH:bool | None=None, # Use two-stage adaptive Benjamini-Hochberg; None infers True for motifs (DAG-grouped families) and False for sequences
    custom_motifs:str | list[str]=[], # Custom motifs if using 'custom' feature set
    transform:str | None=None, # Transformation type: "CLR" or "ALR"; None auto-decides
    gamma:float=0.1, # Uncertainty parameter for CLR transform
    custom_scale:float | dict=0, # Ratio of total signal in group2/group1 for an informed scale model (or group_idx: mean(group)/min(mean(groups)) signal dict for multivariate)
    moderate_variance:bool=True, # Empirical-Bayes variance moderation, with the containment DAG as the prior neighborhood
    glycoproteomics:bool | None=None, # Whether data is from glycoproteomics (rows protein_site_glycan); default: whenever every row label is protein_site_glycan
    level:str='peptide', # Analysis level for glycoproteomics
    monte_carlo:bool=False, # Use Monte Carlo for technical variation
    random_state:int | numpy.random._generator.Generator | None=None, # optional random state for reproducibility
    top_explained:int | None=5, # How many child motifs to name in 'Explained by'; None names all of them
)->glycowork.glycan_data.loader.GlycoDataFrame: # DataFrame with log2FC, p-values, FDR-corrected p-values, and Cohen's d/Mahalanobis distance effect sizes

Performs differential expression analysis using Welch’s t-test (or Hotelling’s T2 for sets) with multiple testing correction on glycomics abundance data

test_df = glycomics_data_loader.human_skin_O_PMC5871710_BCC
print(test_df.groups.mapping, test_df.paired)  # group assignments and pairing come with the dataset
res = get_differential_expression(test_df, motifs = True)
res
{'healthy': ['control_1', 'control_2', 'control_3', 'control_4', 'control_5', 'control_6', 'control_7', 'control_8', 'control_9', 'control_10', 'control_11', 'control_12', 'control_13', 'control_14', 'control_15', 'control_16', 'control_17', 'control_18', 'control_19', 'control_20'], 'cancer': ['tumor_1', 'tumor_2', 'tumor_3', 'tumor_4', 'tumor_5', 'tumor_6', 'tumor_7', 'tumor_8', 'tumor_9', 'tumor_10', 'tumor_11', 'tumor_12', 'tumor_13', 'tumor_14', 'tumor_15', 'tumor_16', 'tumor_17', 'tumor_18', 'tumor_19', 'tumor_20']} True
Chosen grouping: by_motif_family
Glycan Mean abundance Log2FC p-val corr p-val significant corr Levene p-val Effect size Equivalence p-val Explained by Redistribution p-val Residual p-val Residual effect size
12 Neu5Ac(a2-3)Gal 12.345115 0.313599 0.000015 0.000073 True 0.846171 1.286347 1.000000 Disialyl_T_antigen (-0.36) 0.005505 0.000193 1.168046
7 Gal3/6S(b1-3)GalNAc 0.159900 -0.838774 0.000696 0.004176 True 0.846171 -0.847840 1.000000 NaN NaN NaN
0 H_antigen_type2 0.247156 -0.654481 0.001817 0.006209 True 0.996125 -0.788570 1.000000 NaN NaN NaN
1 Internal_LacNAc_type2 2.328696 0.491517 0.003104 0.006209 True 0.608968 0.731626 1.000000 H_antigen_type2 (-1.15) 0.000306 0.000971 0.871276
9 GlcNAc6S(b1-6)GalNAc 1.046247 0.884675 0.004181 0.010466 True 0.996125 0.684443 1.000000 NaN NaN NaN
4 Oglycan_core1 7.956860 0.234965 0.005233 0.010466 True 0.996125 0.692574 1.000000 Disialyl_T_antigen (-0.28) 0.002596 0.000238 1.077778
6 Gal 18.541780 0.184057 0.009787 0.013049 True 0.996125 0.661140 1.000000 Terminal_LacNAc_type2 (-0.60), Internal_LacNAc... 0.002265 1.000000 0.000000
2 Terminal_LacNAc_type2 2.440640 -0.413622 0.014311 0.014311 True 0.608968 -0.577125 1.000000 NaN NaN NaN
10 Neu5Ac 16.553504 0.195722 0.009963 0.021766 True 0.996125 0.666531 1.000000 Neu5Ac(a2-8)Neu5Ac (-0.82), Neu5Ac(a2-6)GalNAc... 0.005061 1.000000 0.000000
14 Neu5Ac(a2-8)Neu5Ac 0.038663 -0.621757 0.013059 0.021766 True 0.608968 -0.582300 1.000000 NaN NaN NaN
8 GalNAc 12.886096 0.121066 0.030163 0.045244 True 0.996125 0.645643 1.000000 Gal3/6S(b1-3)GalNAc (-0.96), GlcNAc6S(b1-6)Gal... 0.003977 1.000000 0.000000
11 Gal(b1-3)GalNAc 12.726196 0.099145 0.071818 0.086182 True 0.842190 0.437094 0.831783 Oglycan_core1 (+0.14), Oglycan_core2 (-0.07) 0.271310 1.000000 0.000000
13 Neu5Ac(a2-6)GalNAc 4.169726 -0.061919 0.493222 0.616527 False 0.996125 -0.152631 0.450845 Disialyl_T_antigen (+0.02) 0.000587 0.000845 -0.913043
3 Disialyl_T_antigen 3.790085 -0.043993 0.654437 0.654437 False 0.996125 -0.099488 0.450845 NaN NaN NaN
5 Oglycan_core2 4.769337 0.025826 0.811772 0.811772 False 0.996125 0.052366 0.450845 NaN NaN NaN

get_volcano

def get_volcano(
    df_res:pandas.DataFrame | str | pathlib.Path, # DataFrame from get_differential_expression with columns [Glycan, Log2FC, p-val, corr p-val]
    y_thresh:float | None=None, # Corrected p threshold for labeling; default: alpha stamped by the analysis function, else 0.05
    x_thresh:float=0, # Absolute x metric threshold for labeling
    n:int | None=None, # Sample size for Bayesian-Adaptive Alpha
    label_changed:bool=True, # Add text labels to significant points
    x_metric:str='Log2FC', # x-axis metric: 'Log2FC' or 'Effect size'; glycosite-level results (glycoproteomics) have no Log2FC and use Effect size
    annotate_volcano:bool=True, # Annotate dots with SNFG images
    filepath:str | pathlib.Path='', # Path to save plot
    title:str | None=None, # Plot title; default: the dataset name the analysis stamped on df_res, '' for none
    **kwargs:Any
)->None: # Displays volcano plot, or returns it as a Jupyter-renderable SVG when annotate_volcano

Creates volcano plot showing -log10(FDR-corrected p-values) vs Log2FC or effect size

get_volcano(res)

get_coverage

def get_coverage(
    df:pandas.DataFrame | str | pathlib.Path, # DataFrame with glycans in rows (col 1), abundances in columns
    filepath:str='', # Path to save plot
    title:str | None=None, # Plot title; None for none, as before
)->None:

Visualizes glycan detection frequency across samples with intensity-based ordering

test_df = test_df.astype({c: float for c in test_df.columns[1:]})
get_coverage(test_df)

get_pca

def get_pca(
    df:pandas.DataFrame | str | pathlib.Path, # DataFrame with glycans in rows (col 1), abundances in columns
    groups:list[int] | pandas.DataFrame | None=None, # Group labels (e.g., [1,1,1,2,2,2,3,3,3]) or metadata DataFrame with 'id' column
    motifs:bool=False, # Analyze motifs instead of sequences
    feature_set:str | list[str]=['known', 'exhaustive'], # Feature sets to use; exhaustive, known, terminal1, terminal2, terminal3, chemical, graph, custom, size_branch
    pc_x:int=1, # Principal component for x-axis
    pc_y:int=2, # Principal component for y-axis
    color:str | None=None, # Column in metadata for color grouping; recommended to be categorical
    shape:str | None=None, # Column in metadata for shape grouping; recommended to be categorical
    size:str | None=None, # Column in metadata for point size control; recommended to be scalar
    filepath:str | pathlib.Path='', # Path to save plot
    custom_motifs:str | list[str]=[], # Custom motifs if using 'custom' feature set
    transform:str | None=None, # Transformation type: "CLR" or "ALR"
    rarity_filter:float=0.05, # Min proportion for non-zero values
    eigenvalues:bool=False, # Plot the explained variance as a separate subplot
    title:str | None=None, # Plot title; None for none, as before
)->None:

Performs PCA on glycan/motif abundance data with group-based visualization

get_pca(test_df, motifs = True, groups = [1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2,1,2])

get_pcoa

def get_pcoa(
    df:pandas.DataFrame | str | pathlib.Path, # Abundance dataframe or filepath (glycans as rows, samples as columns) or a square distance matrix, e.g., from get_distance_matrix or get_biodiversity
    groups:list[str | int] | None=None, # Group label per sample for coloring and PERMANOVA; default: from the frame's contrasts
    dist_func:Union[str, Callable[[list, list], float]]='euclidean', # scipy.spatial.distance metric name or callable on two lists; euclidean on CLR data is the Aitchison distance
    motifs:bool=False, # Analyze motifs instead of sequences
    feature_set:str | list[str]=['known', 'exhaustive'], # Feature sets to use; exhaustive, known, terminal1, terminal2, terminal3, chemical, graph, custom, size_branch
    custom_motifs:str | list[str]=[], # Custom motifs if using 'custom' feature set
    transform:str='CLR', # Transform data before computing distances: 'CLR', 'ALR', or '' for none
    pco_x:int=1, # Principal coordinate for x-axis
    pco_y:int=2, # Principal coordinate for y-axis
    permutations:int=999, # Number of permutations for PERMANOVA
    random_state:int | numpy.random._generator.Generator | None=None, # optional random state for reproducibility
    filepath:str | pathlib.Path='', # Path to save plot
    title:str | None=None, # Plot title; None for none
)->pandas.DataFrame: # Sample coordinates on all principal coordinates, explained variance and PERMANOVA in .attrs

Performs principal coordinate analysis (PCoA) on any sample distance matrix with group-based visualization and PERMANOVA

get_pval_distribution

def get_pval_distribution(
    df_res:pandas.DataFrame | str | pathlib.Path, # Output DataFrame from get_differential_expression
    filepath:str | pathlib.Path='', # Path to save plot
    title:str | None=None, # Plot title; None for none, as before
)->None:

Creates histogram of p-values from differential expression analysis

get_pval_distribution(res)

get_ma

def get_ma(
    df_res:pandas.DataFrame | str | pathlib.Path, # Output DataFrame from get_differential_expression; glycosite-level results are plotted per glycoform
    log2fc_thresh:int=1, # Log2FC threshold for highlighting
    sig_thresh:float | None=None, # Significance threshold for highlighting; defaults to the sample-size-adjusted alpha stored on the results
    filepath:str | pathlib.Path='', # Path to save plot
    title:str | None=None, # Plot title; default: the dataset name the analysis stamped on df_res, '' for none
)->None:

Generates MA plot (mean abundance vs log2 fold change) from differential expression results

get_ma(res)

get_glycanova

def get_glycanova(
    df:pandas.DataFrame | str | pathlib.Path, # DataFrame with glycans in rows (col 1) and abundance values in columns
    groups:list[typing.Any] | None=None, # Group labels for samples (e.g., [1,1,1,2,2,2,3,3,3]); inferred from a GlycoDataFrame's contrasts if omitted
    impute:bool=True, # Replace zeros with impute_biosynthetic predictions
    motifs:bool=False, # Analyze motifs instead of sequences; per glycosite (protein_site_motif) for glycoproteomics data
    feature_set:str | list[str]=['exhaustive', 'known'], # Feature sets to use; exhaustive, known, terminal1, terminal2, terminal3, chemical, graph, custom, size_branch
    min_samples:float=0.1, # Min fraction (0-1) of non-zero samples required
    posthoc:bool=True, # Perform Tukey's HSD test post-hoc
    grouped_BH:bool | None=None, # Use two-stage adaptive Benjamini-Hochberg; None infers True for motifs (DAG-grouped families) and False for sequences
    custom_motifs:str | list[str]=[], # Custom motifs if using 'custom' feature set
    transform:str | None=None, # Transformation type: "CLR" or "ALR"; None auto-decides
    gamma:float=0.1, # Uncertainty parameter for CLR transform
    custom_scale:float | dict=0, # Ratio of total signal in group2/group1 for an informed scale model (or group_idx: mean(group)/min(mean(groups)) signal dict for multivariate)
    moderate_variance:bool=True, # Empirical-Bayes variance moderation, with the containment DAG as the prior neighborhood
    glycoproteomics:bool | None=None, # Whether rows are glycoforms from glycoproteomics instead of glycans; default: whenever every row label is protein_site_glycan
    random_state:int | numpy.random._generator.Generator | None=None, # optional random state for reproducibility
    top_explained:int | None=5, # How many child motifs to name in 'Explained by'; None names all of them
)->tuple[glycowork.glycan_data.loader.GlycoDataFrame, dict[str, pandas.DataFrame]]: # (ANOVA results with F-stats and omega-squared effect sizes, post-hoc results)

Performs one-way ANOVA with omega-squared effect size calculation and optional Tukey’s HSD post-hoc testing on glycomics data across multiple groups

test_df2 = glycomics_data_loader.HIV_gagtransfection_O_PMID35112714

anv, ph = get_glycanova(test_df2, motifs = False)
anv
Glycan F statistic p-val corr p-val significant Effect size
1 Gal(b1-4)GlcNAc(b1-6)[Gal(b1-3)]GalNAc 3.789915 0.042303 0.311002 False 0.221331
3 Neu5Ac(a2-3)Gal(b1-3)[Gal(b1-4)GlcNAc(b1-6)]Ga... 3.111038 0.069111 0.311002 False 0.231571
0 Gal(b1-3)[Neu5Ac(a2-6)]GalNAc 1.749110 0.202223 0.348838 False 0.073247
2 Neu5Ac(a2-3)Gal(b1-3)GalNAc 1.583464 0.232558 0.348838 False 0.201921
4 Neu5Ac(a2-3)Gal(b1-3)[Neu5Ac(a2-6)]GalNAc 1.992994 0.165249 0.348838 False 0.198865
5 Neu5Ac(a2-3)Gal(b1-4)GlcNAc(b1-3/6)[GlcNAc(b1-... 2.111745 0.150018 0.348838 False 0.096276
6 Neu5Ac(a2-3)Gal(b1-4)GlcNAc(b1-6)[Gal(b1-3)]Ga... 0.598796 0.560056 0.605809 False -0.077190
7 Neu5Ac(a2-3)Gal(b1-4)GlcNAc(b1-6)[Neu5Ac(a2-3)... 0.515408 0.605809 0.605809 False -0.056393
8 Neu5Ac(a2-3)Gal(b1-4)GlcNAc6S(b1-6)[Neu5Ac(a2-... 0.731458 0.494973 0.605809 False -0.009179

get_meta_analysis

def get_meta_analysis(
    effect_sizes:numpy.ndarray | list[float], # List of Cohen's d/other effect sizes
    variances:numpy.ndarray | list[float], # Associated variance estimates
    model:str='fixed', # 'fixed' or 'random' effects model
    filepath:str='', # Path to save Forest plot
    study_names:list[str]=[], # Names corresponding to each effect size
    full_output:bool=False, # Return heterogeneity statistics (tau2, Q, I2) and leave-one-out pooling instead of just (effect, p-value)
    title:str | None=None, # Forest plot title; None for none, as before
)->tuple[float, float] | dict: # (combined effect size, two-tailed p-value), or the full result dict when full_output=True

Performs fixed/random effects meta-analysis using DerSimonian-Laird method for between-study variance estimation, with optional Forest plot visualization

get_meta_analysis([-8.759, -6.363, -5.199, -3.952],
                 [7.061, 4.041, 2.919, 1.968])
(-5.326913553837341, 3.0050764424211887e-09)

get_time_series

def get_time_series(
    df:pandas.DataFrame | str | pathlib.Path, # DataFrame with sample IDs as 'sampleID_timepoint_replicate' in col 1 (e.g., T1_h5_r1)
    impute:bool=True, # Replace zeros with impute_biosynthetic predictions
    motifs:bool=False, # Analyze motifs instead of sequences
    feature_set:str | list[str]=['known', 'exhaustive'], # Feature sets to use; exhaustive, known, terminal1, terminal2, terminal3, chemical, graph, custom, size_branch
    degree:int=1, # Polynomial degree for regression
    min_samples:float=0.1, # Min fraction (0-1) of non-zero samples required
    grouped_BH:bool | None=None, # Family-grouped two-stage Benjamini-Hochberg via the motif DAG; None infers True for motifs and False for sequences
    custom_motifs:str | list[str]=[], # Custom motifs if using 'custom' feature set
    transform:str | None=None, # Transformation type: "CLR" or "ALR"; None auto-decides
    gamma:float=0.1, # Uncertainty parameter for CLR transform
    custom_scale:float | dict=0, # Ratio of total signal in group2/group1 for an informed scale model (or group_idx: mean(group)/min(mean(groups)) signal dict for multivariate)
    glycoproteomics:bool | None=None, # Whether rows are glycoforms, ordered by composition containment instead of substructure containment; default: whenever every row label is protein_site_glycan
    random_state:int | numpy.random._generator.Generator | None=None, # optional random state for reproducibility
)->glycowork.glycan_data.loader.GlycoDataFrame: # DataFrame with regression coefficients and FDR-corrected p-values

Analyzes time series glycomics data using polynomial regression

t_dic = {}
t_dic["ID"] = ["D1_h5_r1", "D1_h5_r2", "D1_h5_r3", "D1_h10_r1", "D1_h10_r2", "D1_h10_r3", "D1_h15_r1", "D1_h15_r2", "D1_h15_r3"]
t_dic["Neu5Ac(a2-3)Gal(b1-4)GlcNAc(b1-6)[Gal(b1-3)]GalNAc"] = [0.33, 0.31, 0.35, 1.51, 1.57, 1.66, 2.11, 2.04, 2.09]
t_dic["Fuc(a1-2)Gal(b1-3)GalNAc"] = [0.78, 1.01, 0.98, 0.88, 1.11, 0.72, 1.22, 1.00, 0.54]
t_dic["Neu5Ac(a2-6)GalNAc"] = [0.11, 0.09, 0.14, 0.02, 0.07, 0.10, 0.11, 0.09, 0.08]
get_time_series(pd.DataFrame(t_dic).set_index("ID").T)
Glycan Change p-val corr p-val significant
1 Neu5Ac(a2-3)Gal(b1-4)GlcNAc(b1-6)[Gal(b1-3)]Ga... 0.168078 0.001757 0.005272 True
0 Fuc(a1-2)Gal(b1-3)GalNAc -0.082124 0.012562 0.012562 True
2 Neu5Ac(a2-6)GalNAc -0.095766 0.009956 0.012562 True

get_jtk

def get_jtk(
    df_in:pandas.DataFrame | str | pathlib.Path, # DataFrame with glycans in rows (first column), then groups arranged by ascending timepoints
    timepoints:int, # Number of timepoints (each must have same number of replicates)
    interval:int, # Time units between experimental timepoints
    periods:int | list[int]=[12, 24], # Cycle length(s) to test, in the time units of interval (e.g., hours)
    motifs:bool=False, # Analyze motifs instead of sequences
    feature_set:str | list[str]=['known', 'exhaustive', 'terminal'], # Feature sets to use; exhaustive, known, terminal1, terminal2, terminal3, chemical, graph, custom, size_branch
    custom_motifs:str | list[str]=[], # Custom motifs if using 'custom' feature set
    transform:str | None=None, # Transformation type: "CLR" or "ALR"; None auto-decides
    gamma:float=0.1, # Uncertainty parameter for CLR transform
    correction_method:str='two-stage', # Multiple testing correction method
    grouped_BH:bool | None=None, # Family-grouped two-stage Benjamini-Hochberg via the motif DAG; None infers True for motifs and False for sequences
    glycoproteomics:bool | None=None, # Whether rows are glycoforms, ordered by composition containment instead of substructure containment; default: whenever every row label is protein_site_glycan
    random_state:int | numpy.random._generator.Generator | None=None, # optional random state for reproducibility
)->glycowork.glycan_data.loader.GlycoDataFrame: # DataFrame with JTK results: adjusted p-values, period length, lag phase, amplitude

Identifies rhythmically expressed glycans using Jonckheere-Terpstra-Kendall algorithm for time series analysis

t_dic = {}
t_dic["Neu5Ac(a2-3)Gal(b1-3)GalNAc"] = [0.433138901, 0.149729209, 0.358018822, 0.537641256, 1.526963756, 1.349986672, 0.75156406, 0.736710183]
t_dic["Gal(b1-3)GalNAc"] = [0.919762334, 0.760237184, 0.725566662, 0.459945797, 0.523801515, 0.695106926, 0.627632047, 1.183511209]
t_dic["Gal(b1-3)[Neu5Ac(a2-6)]GalNAc"] = [0.533138901, 0.119729209, 0.458018822, 0.637641256, 1.726963756, 1.249986672, 0.55156406, 0.436710183]
t_dic["Fuc(a1-2)Gal(b1-3)GalNAc"] = [3.862169504, 5.455032837, 3.858163289, 5.614650335, 3.124254095, 4.189550337, 4.641831312, 4.19538484]
tps = 8  # number of timepoints in experiment
periods = [8]  # potential cycles to test
interval = 3  # units of time between experimental timepoints
t_df = pd.DataFrame(t_dic).T
t_df.columns = ["T3", "T6", "T9", "T12", "T15", "T18", "T21", "T24"]
get_jtk(t_df.reset_index(), tps, interval, periods = periods)
Molecule_Name Adjusted_P_value Period_Length Lag_Phase Amplitude significant
0 Neu5Ac(a2-3)Gal(b1-3)GalNAc 0.710523 8 7.0 0.142857 False
1 Gal(b1-3)GalNAc 0.710523 8 6.0 0.285714 False
2 Gal(b1-3)[Neu5Ac(a2-6)]GalNAc 0.710523 8 1.0 0.142857 False
3 Fuc(a1-2)Gal(b1-3)GalNAc 0.710523 8 3.0 0.285714 False 
get_jtk(t_df.reset_index(), tps, interval, periods = periods, motifs = True, feature_set = ['terminal'])
Chosen grouping: by_motif_family
Molecule_Name Adjusted_P_value Period_Length Lag_Phase Amplitude significant
0 Terminal_Fuc(a1-2) 0.173546 8 3.0 0.500000 False
1 Terminal_Gal(b1-3) 0.173546 8 7.0 0.428571 False
2 Terminal_Neu5Ac(a2-3) 0.536187 8 5.0 0.214286 False
3 Terminal_Neu5Ac(a2-3/6) 0.536187 8 5.0 0.285714 False
4 Terminal_Neu5Ac(a2-6) 0.536187 8 1.0 0.357143 False

get_cosinor

def get_cosinor(
    df_in:pandas.DataFrame | str | pathlib.Path, # DataFrame with glycans in rows (first column) and samples in the other columns
    timepoints:int | list[float] | numpy.ndarray, # Number of timepoints, for columns arranged by ascending time with equal replicates (as in get_jtk), or the time of each sample column (any spacing or replication)
    interval:float=1, # Time units between timepoints (only used if timepoints is a number)
    periods:float | list[float]=[24], # Candidate period(s) in time units; each feature gets its best-fitting one, with the p-value Bonferroni-corrected for the number of candidates
    motifs:bool=False, # Analyze motifs instead of sequences
    feature_set:str | list[str]=['known', 'exhaustive', 'terminal'], # Feature sets to use; exhaustive, known, terminal1, terminal2, terminal3, chemical, graph, custom, size_branch
    custom_motifs:str | list[str]=[], # Custom motifs if using 'custom' feature set
    transform:str | None=None, # Transformation type: "CLR" or "ALR"; None auto-decides
    gamma:float=0.1, # Uncertainty parameter for CLR transform
    correction_method:str='two-stage', # Multiple testing correction method
    grouped_BH:bool | None=None, # Family-grouped two-stage Benjamini-Hochberg via the motif DAG; None infers True for motifs and False for sequences
    glycoproteomics:bool | None=None, # Whether rows are glycoforms, ordered by composition containment instead of substructure containment; default: whenever every row label is protein_site_glycan
    random_state:int | numpy.random._generator.Generator | None=None, # optional random state for reproducibility
)->glycowork.glycan_data.loader.GlycoDataFrame: # DataFrame with cosinor results: best period, mesor, amplitude, acrophase (peak time, on the time axis of timepoints), R2, raw and adjusted p-values

Identifies rhythmically expressed glycans or motifs with single-harmonic cosinor regression, reporting amplitude and acrophase (peak time) next to an F-test against no rhythm; unlike get_jtk, timepoints may be unevenly spaced or unevenly replicated

get_cosinor(t_df.reset_index(), tps, interval, periods = [12, 24])
Molecule_Name P_value Period_Length Mesor Amplitude Acrophase R2 Adjusted_P_value significant
0 Neu5Ac(a2-3)Gal(b1-3)GalNAc 0.005179 24 -0.750058 0.636106 14.912602 0.907681 0.010358 True
1 Gal(b1-3)GalNAc 0.010804 24 -0.612597 0.644458 0.958050 0.876116 0.010804 True
2 Gal(b1-3)[Neu5Ac(a2-6)]GalNAc 0.069740 24 -0.805610 0.527460 12.003714 0.738795 0.069740 True
3 Fuc(a1-2)Gal(b1-3)GalNAc 0.318671 24 2.072303 0.388154 1.998049 0.520349 0.318671 False

get_biodiversity

def get_biodiversity(
    df:pandas.DataFrame | str | pathlib.Path, # DataFrame with glycans in rows (col 1), abundances in columns
    group1:list[str | int] | None=None, # First group column indices or group labels; default: from the frame's contrasts
    group2:list[str | int] | None=None, # Second group indices or additional group labels; default: from the frame's contrasts
    metrics:list[str]=['alpha', 'beta'], # Diversity metrics to calculate
    motifs:bool=False, # Analyze motifs instead of sequences
    feature_set:str | list[str]=['exhaustive', 'known'], # Feature sets to use; exhaustive, known, terminal1, terminal2, terminal3, chemical, graph, custom, size_branch
    custom_motifs:str | list[str]=[], # Custom motifs if using 'custom' feature set
    paired:bool | None=None, # Whether samples are paired; default: from the frame
    permutations:int=999, # Number of permutations for ANOSIM/PERMANOVA
    transform:str | None=None, # Transformation type: "CLR" or "ALR"
    dist_func:Union[str, Callable, NoneType]=None, # scipy.spatial.distance metric name or callable for beta diversity; default: braycurtis if transform = "Nothing", else euclidean (Aitchison distance on log-ratios)
    gamma:float=0.1, # Uncertainty parameter for CLR transform
    custom_scale:float | dict=0, # Ratio of total signal in group2/group1 for an informed scale model (or group_idx: mean(group)/min(mean(groups)) signal dict for multivariate)
    random_state:int | numpy.random._generator.Generator | None=None, # optional random state for reproducibility
    circadian:bool=False, # test whether diversity changes rhythmically over time via JTK
    timepoints:int | None=None, # number of timepoints, columns ordered by ascending timepoint (required if circadian)
    interval:int=1, # time units between timepoints (only relevant if circadian)
    periods:int | list[int]=[12, 24], # cycle length(s) to test (only relevant if circadian)
    glycoproteomics:bool | None=None, # Whether rows are glycoforms (protein_site_glycan): alpha diversity becomes each glycosite's microheterogeneity (rows protein_site_metric) and beta diversity the Aitchison distance over glycosites; default: whenever every row label is protein_site_glycan
)->tuple: # First DataFrame with diversity indices and test statistics, second with beta-diversity distance matrix

Calculates alpha (Shannon/Simpson) and beta (ANOSIM/PERMANOVA) diversity measures from glycomics data, or per-glycosite microheterogeneity from glycoproteomics data

res = get_biodiversity(test_df, motifs = True)
res
(                       Metric  Group1 mean  Group2 mean     p-val  \
 0           simpson_diversity     0.876756     0.874348  0.000443   
 1           shannon_diversity     2.244523     2.225758  0.001255   
 2     Beta diversity (ANOSIM)          NaN          NaN  0.002000   
 3  Beta diversity (PERMANOVA)          NaN          NaN  0.002000   
 4            species_richness    15.000000    15.000000  1.000000   
 
    Effect size  corr p-val  significant  
 0    -0.948203    0.000443         True  
 1    -0.846077    0.001255         True  
 2     0.188566    0.002000         True  
 3     5.583354    0.002000         True  
 4     0.000000    1.000000        False  ,
             control_1  control_2  control_3  control_4  control_5  control_6  \
 control_1    0.000000   2.468605   3.219773   3.547879   1.683100   3.475447   
 control_2    2.468605   0.000000   2.297694   2.679064   2.384198   2.844599   
 control_3    3.219773   2.297694   0.000000   1.322106   3.144578   4.631515   
 control_4    3.547879   2.679064   1.322106   0.000000   3.487061   4.948857   
 control_5    1.683100   2.384198   3.144578   3.487061   0.000000   3.180142   
 control_6    3.475447   2.844599   4.631515   4.948857   3.180142   0.000000   
 control_7    1.705143   2.441621   3.493949   3.690858   2.772165   3.266817   
 control_8    1.988837   1.877254   3.276978   3.497749   2.487358   2.458114   
 control_9    2.910796   2.807673   3.864873   4.336024   3.615115   3.444445   
 control_10   1.950670   2.641473   4.123190   4.470577   2.336412   2.047429   
 control_11   2.126936   1.983422   2.917195   3.538471   2.232348   2.648045   
 control_12   3.890187   2.235635   2.502925   3.435423   3.788631   4.262512   
 control_13   3.262471   1.947637   3.633341   4.139216   3.772189   3.181875   
 control_14   3.204359   1.411597   2.634386   3.347329   3.349799   3.493967   
 control_15   3.469505   1.462537   3.287079   3.460895   3.470620   2.860301   
 control_16   3.134200   1.686293   2.090097   2.971119   3.254491   4.000192   
 control_17   2.940751   1.649726   2.610784   3.345235   3.201940   3.509533   
 control_18   2.715693   2.738157   1.968326   1.598788   3.212088   5.065846   
 control_19   3.911231   5.550462   5.826237   5.787237   4.783100   5.446369   
 control_20   3.582369   3.552533   2.318463   2.968550   3.928537   5.779615   
 tumor_1      2.288187   2.491535   2.427182   3.491011   2.324187   3.931309   
 tumor_2      2.491034   3.566609   4.186507   4.740574   2.689456   3.588158   
 tumor_3      3.838434   4.443823   4.796325   5.851548   4.357685   4.660899   
 tumor_4      2.892877   4.526809   4.582073   5.296299   3.482623   5.023658   
 tumor_5      4.379914   4.904538   5.884048   6.692547   4.903481   4.287772   
 tumor_6      4.546253   5.346522   6.407226   7.084310   4.991457   4.204977   
 tumor_7      3.912857   3.276828   3.804088   4.855699   3.854233   3.697074   
 tumor_8      3.285207   3.995706   4.839851   5.410133   4.110546   3.791720   
 tumor_9      4.169500   4.544534   5.086052   5.739536   5.054963   4.893838   
 tumor_10     3.344534   3.692349   4.949658   5.702753   3.761015   2.898755   
 tumor_11     3.020348   4.115141   4.495357   5.155513   3.211059   4.023108   
 tumor_12     3.744053   2.348433   3.020770   3.821250   3.996783   4.084373   
 tumor_13     2.705654   1.134934   3.095735   3.658740   2.715985   2.524830   
 tumor_14     2.641863   1.225830   2.200278   2.775819   2.841152   3.523533   
 tumor_15     3.960528   3.186287   3.286791   4.202841   4.580778   5.060585   
 tumor_16     4.113015   3.096309   3.603232   4.593119   4.490088   4.432897   
 tumor_17     2.907768   1.297669   2.070248   2.650838   3.106771   3.856230   
 tumor_18     2.938670   2.967581   3.988129   4.690057   2.940891   3.068111   
 tumor_19     4.395063   6.188330   6.388777   6.233811   5.277394   6.220262   
 tumor_20     3.724356   3.180760   2.196601   3.356134   4.002378   5.287895   
 
             control_7  control_8  control_9  control_10  ...  tumor_11  \
 control_1    1.705143   1.988837   2.910796    1.950670  ...  3.020348   
 control_2    2.441621   1.877254   2.807673    2.641473  ...  4.115141   
 control_3    3.493949   3.276978   3.864873    4.123190  ...  4.495357   
 control_4    3.690858   3.497749   4.336024    4.470577  ...  5.155513   
 control_5    2.772165   2.487358   3.615115    2.336412  ...  3.211059   
 control_6    3.266817   2.458114   3.444445    2.047429  ...  4.023108   
 control_7    0.000000   1.201477   1.948433    1.758108  ...  3.405495   
 control_8    1.201477   0.000000   1.760593    1.366530  ...  3.217559   
 control_9    1.948433   1.760593   0.000000    2.272090  ...  2.993728   
 control_10   1.758108   1.366530   2.272090    0.000000  ...  2.688525   
 control_11   2.048773   1.443322   1.767267    1.677971  ...  2.281849   
 control_12   4.077566   3.598668   3.623234    4.208491  ...  4.621330   
 control_13   2.924184   2.591659   2.846241    3.119307  ...  4.721637   
 control_14   3.121872   2.648195   2.831547    3.352762  ...  4.398215   
 control_15   3.066343   2.304979   2.871188    3.130537  ...  4.738938   
 control_16   3.363026   3.078532   3.430205    3.742200  ...  4.556415   
 control_17   2.865466   2.408771   2.301380    3.043546  ...  3.735184   
 control_18   3.119995   3.285354   4.114189    4.130747  ...  4.935647   
 control_19   3.402467   4.034578   4.411020    3.879292  ...  4.231050   
 control_20   4.384365   4.421777   4.835912    4.986077  ...  5.012777   
 tumor_1      3.088164   2.958317   3.303680    3.103777  ...  2.989727   
 tumor_2      2.886284   2.642746   2.619393    2.121498  ...  0.850462   
 tumor_3      4.119708   4.104650   3.544329    3.776558  ...  2.707058   
 tumor_4      3.680360   3.854386   3.714950    3.427984  ...  1.770596   
 tumor_5      4.272613   4.046936   3.199465    3.451164  ...  2.659752   
 tumor_6      4.423976   4.213954   3.607641    3.377535  ...  2.632130   
 tumor_7      3.856605   3.465342   3.035033    3.515631  ...  3.168408   
 tumor_8      2.380499   2.663546   1.935332    2.468447  ...  2.724558   
 tumor_9      3.385319   3.572012   2.078336    3.675380  ...  3.171551   
 tumor_10     3.207591   2.889885   2.536034    2.272368  ...  2.505579   
 tumor_11     3.405495   3.217559   2.993728    2.688525  ...  0.000000   
 tumor_12     3.767588   3.261217   3.055137    3.879619  ...  4.379136   
 tumor_13     2.738463   2.122849   2.803993    2.548079  ...  4.101109   
 tumor_14     2.598313   2.097768   2.364797    2.950728  ...  3.808831   
 tumor_15     3.934628   3.824875   3.281342    4.426797  ...  4.567879   
 tumor_16     4.092128   3.724746   3.219082    4.149805  ...  4.318607   
 tumor_17     2.955076   2.593234   3.021394    3.437302  ...  4.472500   
 tumor_18     3.135558   2.523906   2.241670    2.250767  ...  1.951492   
 tumor_19     4.080270   4.697457   5.039276    4.539426  ...  4.678628   
 tumor_20     4.132016   4.027648   3.950113    4.613526  ...  4.351279   
 
             tumor_12  tumor_13  tumor_14  tumor_15  tumor_16  tumor_17  \
 control_1   3.744053  2.705654  2.641863  3.960528  4.113015  2.907768   
 control_2   2.348433  1.134934  1.225830  3.186287  3.096309  1.297669   
 control_3   3.020770  3.095735  2.200278  3.286791  3.603232  2.070248   
 control_4   3.821250  3.658740  2.775819  4.202841  4.593119  2.650838   
 control_5   3.996783  2.715985  2.841152  4.580778  4.490088  3.106771   
 control_6   4.084373  2.524830  3.523533  5.060585  4.432897  3.856230   
 control_7   3.767588  2.738463  2.598313  3.934628  4.092128  2.955076   
 control_8   3.261217  2.122849  2.097768  3.824875  3.724746  2.593234   
 control_9   3.055137  2.803993  2.364797  3.281342  3.219082  3.021394   
 control_10  3.879619  2.548079  2.950728  4.426797  4.149805  3.437302   
 control_11  2.778048  2.123021  1.861858  3.332524  3.067669  2.508273   
 control_12  1.203142  2.195039  1.751476  1.988133  1.777293  1.630189   
 control_13  1.996080  1.343739  2.001572  2.643974  2.466764  1.896976   
 control_14  1.185084  1.199938  1.102742  2.099911  1.929396  0.967147   
 control_15  2.345477  1.527028  1.745469  3.469990  3.215957  1.915264   
 control_16  1.637987  1.727284  1.494080  2.062082  2.152106  0.978943   
 control_17  1.135477  1.559092  0.860537  1.912926  1.780359  1.247750   
 control_18  3.818196  3.532707  2.715008  3.907790  4.509003  2.519622   
 control_19  6.805516  5.945286  5.676018  6.763438  6.955198  6.117307   
 control_20  3.389918  3.867231  3.142313  3.079027  3.701748  2.784954   
 tumor_1     2.816145  2.557554  2.348743  2.912569  2.874042  2.480550   
 tumor_2     4.019861  3.566406  3.295896  4.326440  4.118325  3.963429   
 tumor_3     3.838788  4.034084  4.036864  3.520094  3.152610  4.415768   
 tumor_4     4.610573  4.489025  4.101568  4.376777  4.434165  4.571735   
 tumor_5     4.356852  4.370615  4.490635  4.432914  3.875373  5.072272   
 tumor_6     5.233723  4.887966  5.100473  5.426741  4.882806  5.739144   
 tumor_7     2.914164  3.014374  3.072493  3.222631  2.532553  3.465164   
 tumor_8     4.434114  3.908547  3.844830  4.375099  4.234550  4.395062   
 tumor_9     4.212275  4.469594  3.995506  3.892852  3.877787  4.597440   
 tumor_10    3.683185  3.120813  3.583792  4.029954  3.422198  4.099940   
 tumor_11    4.379136  4.101109  3.808831  4.567879  4.318607  4.472500   
 tumor_12    0.000000  2.047845  1.573929  1.486088  1.187239  1.651741   
 tumor_13    2.047845  0.000000  1.565690  2.983117  2.623838  1.573725   
 tumor_14    1.573929  1.565690  0.000000  2.293256  2.375453  0.925990   
 tumor_15    1.486088  2.983117  2.293256  0.000000  1.205507  2.257323   
 tumor_16    1.187239  2.623838  2.375453  1.205507  0.000000  2.459471   
 tumor_17    1.651741  1.573725  0.925990  2.257323  2.459471  0.000000   
 tumor_18    2.917487  2.704761  2.570243  3.540438  3.050259  3.275947   
 tumor_19    7.376543  6.607802  6.213383  7.300621  7.581433  6.668810   
 tumor_20    2.339844  3.342930  2.519523  1.852860  2.337824  2.390821   
 
             tumor_18  tumor_19  tumor_20  
 control_1   2.938670  4.395063  3.724356  
 control_2   2.967581  6.188330  3.180760  
 control_3   3.988129  6.388777  2.196601  
 control_4   4.690057  6.233811  3.356134  
 control_5   2.940891  5.277394  4.002378  
 control_6   3.068111  6.220262  5.287895  
 control_7   3.135558  4.080270  4.132016  
 control_8   2.523906  4.697457  4.027648  
 control_9   2.241670  5.039276  3.950113  
 control_10  2.250767  4.539426  4.613526  
 control_11  1.494133  5.163824  3.312187  
 control_12  3.322332  7.692524  2.010076  
 control_13  3.321105  6.858445  3.569878  
 control_14  2.972485  7.008998  2.609105  
 control_15  3.244148  6.827899  4.004887  
 control_16  3.453214  7.081476  1.956607  
 control_17  2.342762  6.526540  2.466054  
 control_18  4.577234  5.604184  3.267136  
 control_19  5.286767  1.240709  6.581604  
 control_20  4.650972  6.962530  1.896271  
 tumor_1     2.678595  6.013224  2.053539  
 tumor_2     1.574194  4.531684  4.207256  
 tumor_3     2.890704  6.358839  3.609429  
 tumor_4     2.938220  4.706202  4.011672  
 tumor_5     2.494305  6.177002  5.085724  
 tumor_6     2.961191  5.714909  5.926882  
 tumor_7     2.447520  6.923979  3.227349  
 tumor_8     2.999749  4.377423  4.774996  
 tumor_9     3.236675  5.257604  4.588223  
 tumor_10    1.958069  5.821096  4.533396  
 tumor_11    1.951492  4.678628  4.351279  
 tumor_12    2.917487  7.376543  2.339844  
 tumor_13    2.704761  6.607802  3.342930  
 tumor_14    2.570243  6.213383  2.519523  
 tumor_15    3.540438  7.300621  1.852860  
 tumor_16    3.050259  7.581433  2.337824  
 tumor_17    3.275947  6.668810  2.390821  
 tumor_18    0.000000  5.799777  3.789222  
 tumor_19    5.799777  0.000000  7.117017  
 tumor_20    3.789222  7.117017  0.000000  
 
 [40 rows x 40 columns])

get_SparCC

def get_SparCC(
    df1:pandas.DataFrame | str | pathlib.Path, # First DataFrame with glycans in rows (col 1) and abundances in columns
    df2:pandas.DataFrame | str | pathlib.Path, # Second DataFrame with same format as df1
    motifs:bool=False, # Analyze motifs instead of sequences
    feature_set:str | list[str]=['known', 'exhaustive'], # Feature sets to use; exhaustive, known, terminal1, terminal2, terminal3, chemical, graph, custom, size_branch
    custom_motifs:str | list[str]=[], # Custom motifs if using 'custom' feature set
    transform:str | None=None, # Transformation type: "CLR" or "ALR"
    gamma:float=0.1, # Uncertainty parameter for CLR transform
    partial_correlations:bool=False, # Use regularized partial correlations
    random_state:int | numpy.random._generator.Generator | None=None, # optional random state for reproducibility
)->tuple[pandas.DataFrame, pandas.DataFrame]: # (Spearman correlation matrix, FDR-corrected p-value matrix)

Calculates SparCC (Sparse Correlations for Compositional Data) between two matching datasets (e.g., glycomics)

df1 = glycomics_data_loader.time_series_N_PMID32149347
df2 = glycomics_data_loader.time_series_O_PMID32149347
df1 = pd.merge(df1, df2[['ID']], on = 'ID', how = 'inner')
df2 = pd.merge(df2, df1[['ID']], on = 'ID', how = 'inner')
df1 = df1.set_index(df1.columns.tolist()[0]).T.reset_index()
df2 = df2.set_index(df2.columns.tolist()[0]).T.reset_index()

corr, pval = get_SparCC(df1, df2, motifs = True, transform = "CLR")
import seaborn as sns
sns.clustermap(corr)

get_roc

def get_roc(
    df:pandas.DataFrame | str | pathlib.Path, # DataFrame with glycans in rows (col 1), abundances in columns
    group1:list[str | int] | None=None, # First group indices/names; default: from the frame's contrasts
    group2:list[str | int] | None=None, # Second group indices/names; default: from the frame's contrasts
    motifs:bool=False, # Analyze motifs instead of sequences
    feature_set:str | list[str]=['known', 'exhaustive'], # Feature sets to use; exhaustive, known, terminal1, terminal2, terminal3, chemical, graph, custom, size_branch
    paired:bool | None=None, # Whether samples are paired; default: from the frame
    impute:bool=True, # Replace zeros with impute_biosynthetic predictions
    min_samples:float=0.1, # Min fraction (0-1) of non-zero samples required
    custom_motifs:str | list[str]=[], # Custom motifs if using 'custom' feature set
    transform:str | None=None, # Transformation type: "CLR" or "ALR"
    gamma:float=0.1, # Uncertainty parameter for CLR transform
    custom_scale:float | dict=0, # Ratio of total signal in group2/group1 for an informed scale model (or group_idx: mean(group)/min(mean(groups)) signal dict for multivariate)
    filepath:str | pathlib.Path='', # Path to save ROC plot
    multi_score:bool=False, # Find best multi-glycan score
    random_state:int | numpy.random._generator.Generator | None=None, # optional random state for reproducibility
    title:str | None=None, # Plot title; None keeps the default, '' removes it
)->list[tuple[str, float]] | dict[typing.Any, tuple[str, float]] | tuple['LogisticRegression', float, list[str]]: # (feature, ROC AUC) pairs sorted by AUC; with more than two groups, class: (best feature, AUC); with multi_score, (model, ROC AUC, selected features)

Calculates ROC curves and AUC scores for glycans/motifs or multi-glycan classifiers

get_roc(test_df, motifs = True)

[('Neu5Ac(a2-3)Gal', 0.82),
 ('Neu5Ac', 0.7675000000000001),
 ('Gal(b1-3)GalNAc', 0.7525),
 ('GlcNAc6S(b1-6)GalNAc', 0.7225),
 ('Internal_LacNAc_type2', 0.6975),
 ('Gal', 0.6699999999999999),
 ('Oglycan_core1', 0.655),
 ('GalNAc', 0.6075),
 ('Oglycan_core2', 0.54),
 ('Neu5Ac(a2-6)GalNAc', 0.485),
 ('Disialyl_T_antigen', 0.48),
 ('Neu5Ac(a2-8)Neu5Ac', 0.375),
 ('Terminal_LacNAc_type2', 0.29),
 ('H_antigen_type2', 0.275),
 ('Gal3/6S(b1-3)GalNAc', 0.2225)]

get_distance_matrix

def get_distance_matrix(
    df:pandas.DataFrame | str | pathlib.Path, # Input dataframe or filepath (.csv/.tsv/.xlsx), glycans as rows and samples as columns
    dist_func:Union[str, Callable[[list, list], float]]='euclidean', # scipy.spatial.distance metric name or callable on two lists; euclidean on CLR data is the Aitchison distance
    compare:str='samples', # What to compare pairwise: 'samples' or 'features' (glycans/motifs)
    motifs:bool=False, # Analyze motifs instead of sequences
    feature_set:str | list[str]=['known'], # Feature sets to use; exhaustive, known, terminal1, terminal2, terminal3, chemical, graph, custom, size_branch
    custom_motifs:str | list[str]=[], # Custom motifs if using 'custom' feature set
    transform:str | None='CLR', # Transform data before comparing: 'CLR', 'ALR', or None
    index_col:str='glycan', # Column to use as index
)->pandas.DataFrame: # Square distance matrix, e.g., for dendrogram_from_distance

Calculates pairwise distances between samples or glycans/motifs from an abundance dataframe, with optional motif quantification and compositional transformation

get_lectin_array

def get_lectin_array(
    df:pandas.DataFrame | str | pathlib.Path, # DataFrame with samples as rows and lectins as columns, first column containing sample IDs
    group1:list[str | int] | None=None, # First group sample names or row indices (0-based, or 1-based when no index is 0), or one group label per sample if group2 is omitted; inferred from a GlycoDataFrame's contrasts if omitted
    group2:list[str | int] | None=None, # Second group sample names or row indices; inferred from a GlycoDataFrame's contrasts if omitted
    paired:bool | None=None, # Whether samples are paired; inferred from a GlycoDataFrame if omitted
    transform:str='', # Optional log2 transformation
)->pandas.DataFrame: # DataFrame with altered glycan motifs, supporting lectins, and effect sizes

Analyzes lectin microarray data by mapping lectin binding patterns to glycan motifs, calculating Cohen’s d effect sizes between groups and clustering results by significance

lectin_df = lectin_array_data_loader.A549_influenza_PMID33046650
get_lectin_array(lectin_df, [5,6,7], [8,9,10])
Lectin "Ab-LeB-1" is not found in our annotated lectin library and is excluded from analysis.
Lectin "APA" is not found in our annotated lectin library and is excluded from analysis.
Lectin "APP" is not found in our annotated lectin library and is excluded from analysis.
Lectin "Blood Group B [CLCP-19B]" is not found in our annotated lectin library and is excluded from analysis.
Lectin "Blood Group H2" is not found in our annotated lectin library and is excluded from analysis.
Lectin "CA19-9 [121SLE]" is not found in our annotated lectin library and is excluded from analysis.
Lectin "CCA" is not found in our annotated lectin library and is excluded from analysis.
Lectin "CD15 [ICRF29-2]" is not found in our annotated lectin library and is excluded from analysis.
Lectin "CD15 [MY-1]" is not found in our annotated lectin library and is excluded from analysis.
Lectin "CD15 [SP-159]" is not found in our annotated lectin library and is excluded from analysis.
Lectin "Forssman" is not found in our annotated lectin library and is excluded from analysis.
Lectin "IAA" is not found in our annotated lectin library and is excluded from analysis.
Lectin "IRA" is not found in our annotated lectin library and is excluded from analysis.
Lectin "Le X [P12]" is not found in our annotated lectin library and is excluded from analysis.
Lectin "Lewis A [7LE]" is not found in our annotated lectin library and is excluded from analysis.
Lectin "Lewis B [218]" is not found in our annotated lectin library and is excluded from analysis.
Lectin "Lewis Y [F3]" is not found in our annotated lectin library and is excluded from analysis.
Lectin "LFA" is not found in our annotated lectin library and is excluded from analysis.
Lectin "LPA" is not found in our annotated lectin library and is excluded from analysis.
Lectin "MNA-M " is not found in our annotated lectin library and is excluded from analysis.
Lectin "MUC5Ac Ab" is not found in our annotated lectin library and is excluded from analysis.
Lectin "PMA" is not found in our annotated lectin library and is excluded from analysis.
Lectin "PTA_1" is not found in our annotated lectin library and is excluded from analysis.
Lectin "PTA_2" is not found in our annotated lectin library and is excluded from analysis.
Lectin "SNA-S" is not found in our annotated lectin library and is excluded from analysis.
Lectin "SNA-V" is not found in our annotated lectin library and is excluded from analysis.
Lectin "VFA" is not found in our annotated lectin library and is excluded from analysis.
motif named_motifs lectin(s) change score significance
38 Neu5Ac(a2-6)Gal(b1-4)GlcNAc [Internal_LacNAc_type2] PSL, SNA, TJA-I, WGA_1, WGA_2 down 11.34 highly significant
39 Neu5Ac(a2-6)Gal(b1-3)GlcNAc [Internal_LacNAc_type1] PSL, SNA, TJA-I, WGA_1, WGA_2 down 11.34 highly significant
7 Man(a1-2) [] ASA, Con A, CVN, HHL, SVN_1, GRFT, SVN_2, SNA-... up 4.83 moderately significant
14 Gal(b1-4)GlcNAc(b1-2)Man(a1-3)[Gal(b1-4)GlcNAc... [Chitobiose, Trimannosylcore, Terminal_LacNAc_... CA, CAA, DSA_1, DSA_2, DSA_3, AMA, BDA, BPA, C... up 3.51 moderately significant
4 Gal(b1-3)GalNAc [Oglycan_core1] ACA, AIA, MPA, PNA_1, PNA_2, BDA, BPA up 3.48 moderately significant
43 Neu5Ac(a2-6)GalNAc(b1-4)GlcNAc [Internal_LacdiNAc_type2] SNA, WGA_1, WGA_2 down 2.99 moderately significant
10 Gal(b1-4)GlcNAc(b1-2)Man(a1-3)[GlcNAc(b1-4)][G... [Chitobiose, Trimannosylcore, Terminal_LacNAc_... Blackbean, Calsepa, PHA-E_1, PHA-E_2, AMA, BDA... up 2.66 moderately significant
47 GlcNAc(b1-2)Man(a1-3)[GlcNAc(b1-2)Man(a1-6)]Ma... [Chitobiose, Trimannosylcore, core_fucose, Ngl... TL, AAL, AMA, AOL, Con A, GNA, GNL, HHL, LcH, ... up 2.36 moderately significant
18 Man(a1-6) [] Con A, GNA, GNL, HHL, NPA, SNA-II, UDA up 2.30 moderately significant
17 Man(a1-3) [] Con A, GNA, GNL, HHL, NPA, SNA-II, UDA up 2.30 moderately significant
22 Gal(b1-4)GlcNAc(b1-2)[Gal(b1-4)GlcNAc(b1-4)]Ma... [Chitobiose, Trimannosylcore, Terminal_LacNAc_... DSA_1, DSA_2, DSA_3, AMA, BDA, Blackbean, BPA,... up 2.05 moderately significant
3 Fuc(a1-6) [] AAL, AOL, LcH, PSA up 1.70 moderately significant
6 Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc [Chitobiose, Trimannosylcore] AMA, Con A, GNA, GNL, HHL, NPA, SNA-II, UDA up 1.58 moderately significant
11 GlcNAc(b1-2)Man(a1-3)[GlcNAc(b1-2)[GlcNAc(b1-6... [Chitobiose, Trimannosylcore, Nglycan_complex] Blackbean, PHA-L, AMA, Con A, GNA, GNL, HHL, N... up 1.45 low significance
42 GlcNAc(b1-2)[GlcNAc(b1-6)]Man(a1-6)[GlcNAc(b1-... [Chitobiose, Trimannosylcore, bisectingGlcNAc,... RPA, AMA, Blackbean, Con A, GNA, GNL, HHL, NPA... up 1.40 low significance
41 GlcNAc(b1-2)[GlcNAc(b1-4)]Man(a1-3)[GlcNAc(b1-... [Chitobiose, Trimannosylcore, bisectingGlcNAc,... RPA, AMA, Con A, GNA, GNL, HHL, NPA, SNA-II, U... up 1.36 low significance
15 Gal(b1-3)GalNAc(b1-4)[Neu5Ac(a2-3)]Gal(b1-4)Gl... [Internal_LacNAc_type2] Cholera Toxin, ACA, AIA, BDA, BPA, MPA, MAA_1,... up 1.26 low significance
23 Gal(b1-4)GlcNAc [Terminal_LacNAc_type2] ECA, RCA120, Ricin B Chain, SJA, BDA, BPA up 1.05 low significance
5 GlcNAc(b1-3)GalNAc [Oglycan_core3] AIA, UEA-II, WGA_1, WGA_2 up 0.86 low significance
27 Gal(a1-4) [] GS-I_1, GS-I_2, GS-I_3, GS-I_4, MNA-G, PA-IL up 0.83 low significance
26 Gal(a1-3) [] GS-I_1, GS-I_2, GS-I_3, GS-I_4, MNA-G, PA-IL up 0.83 low significance
30 Gal(b1-4)GlcNAc(b1-3) [Terminal_LacNAc_type2] LEA_1, LEA_2, STA, BDA, BPA, ECA, RCA120, Rici... up 0.55 low significance
25 Gal(a1-3)Gal [] EEA, EEL, MOA, GS-I_1, GS-I_2, GS-I_3, GS-I_4,... up 0.51 low significance
37 Gal(a1-3)GalNAc [] MOA, GS-I_1, GS-I_2, GS-I_3, GS-I_4, MNA-G, PA-IL up 0.46 low significance
40 Fuc(a1-2)Gal(b1-4)GlcNAc [H_antigen_type2, Internal_LacNAc_type2] PTL-II, TJA-II, UEA-I, UEA-II, AAA, AAL, AOL down 0.41 low significance
20 GalNAc(a1-4) [] GHA, HAA, HPA, CSA, GS-I_1, GS-I_2, GS-I_3, GS... up 0.39 low significance
19 GalNAc(a1-3) [] GHA, HAA, HPA, CSA, GS-I_1, GS-I_2, GS-I_3, GS... up 0.39 low significance
21 GalNAc(a1-3)GalNAc(b1-3) [] DBA, SBA, GHA, HAA, HPA, CSA, GS-I_1, GS-I_2, ... up 0.21 low significance
46 Fuc(a1-2)Gal(b1-3)GalNAc [H_type3, Oglycan_core1] TJA-II, AAA, AAL, AOL up 0.21 low significance
24 GalNAc(b1-4)GlcNAc [Terminal_LacdiNAc_type2] ECA, STA, CSA, SBA, VVA_1, VVA_2, WFA, BPA, WG... up 0.20 low significance
16 Fuc(a1-2)Gal(b1-3)GalNAc(b1-4)[Neu5Ac(a2-3)]Ga... [Internal_LacNAc_type2, H_type3] Cholera Toxin, AAA, AAL, AOL, TJA-II, MAA_1, M... up 0.18 low significance
34 Neu5Ac(a2-3)Gal(b1-3)GalNAc [Oglycan_core1] MAL-II, WGA_1, WGA_2 down 0.16 low significance
13 GalNAc(b1-4) [] CSA, SBA, VVA_1, VVA_2, WFA, BPA, WGA_1, WGA_2 up 0.13 low significance
12 GalNAc(b1-3) [] CSA, SBA, VVA_1, VVA_2, WFA, BPA, WGA_1, WGA_2 up 0.13 low significance
32 Gal3S(b1-4)GlcNAc [] MAA_1, MAA_2, MAL-I, MAL-II down 0.12 low significance
28 GlcNAc(a1-3) [] HAA, HPA, WGA_1, WGA_2 up 0.12 low significance
29 GlcNAc(a1-4) [] HAA, HPA, WGA_1, WGA_2 up 0.12 low significance
0 Fuc(a1-2) [] AAA, AAL, AOL up 0.09 low significance
36 Gal3S(b1-4) [] MAL-II down 0.08 low significance
35 Gal3S(b1-3) [] MAL-II down 0.08 low significance
44 Fuc(a1-2)Gal(b1-4)GalNAc(b1-3) [] SNA-II, AAA, AAL, AOL, UEA-II up 0.08 low significance
9 Gal(b1-4) [] BDA, BPA up 0.05 low significance
8 Gal(b1-3) [] BDA, BPA up 0.05 low significance
49 Fuc(a1-2)Gal(b1-4)GalNAc [] UEA-II, AAA, AAL, AOL up 0.05 low significance
33 Neu5Ac(a2-3)Gal(b1-4)GlcNAc [Internal_LacNAc_type2] MAA_1, MAA_2, MAL-I, WGA_1, WGA_2 down 0.04 low significance
2 Fuc(a1-4) [] AAL, AOL down 0.03 low significance
1 Fuc(a1-3) [] AAL, AOL, Lotus down 0.03 low significance
31 GlcNAc(b1-4)GlcNAc(b1-4) [Chitobiose] LEA_1, LEA_2, WGA_1, WGA_2 down 0.01 low significance
50 GlcNAc(b1-3) [] WGA_1, WGA_2 down 0.01 low significance
51 GlcNAc(b1-4) [] WGA_1, WGA_2 down 0.01 low significance
45 GlcNAc(b1-4)GlcNAc(b1-4)GlcNAc(b1-4) [Chitobiose] STA, LEA_1, LEA_2, WGA_1, WGA_2 down 0.00 low significance
48 GlcNAc(b1-3)Gal [] UEA-II, WGA_1, WGA_2 up 0.00 low significance
52 Neu5Ac(a2-3) [] WGA_1, WGA_2 down 0.00 low significance
53 Neu5Ac(a2-6) [] WGA_1, WGA_2 down 0.00 low significance
54 Neu5Ac(a2-8) [] WGA_1, WGA_2 down 0.00 low significance

get_glycoshift_per_site

def get_glycoshift_per_site(
    df:pandas.DataFrame | str | pathlib.Path, # DataFrame with rows formatted as 'protein_site_composition' in col 1, abundances in remaining cols
    group1:list[str | int] | None=None, # First group indices/names or group labels for multi-group; default: from the frame's contrasts
    group2:list[str | int] | None=None, # Second group indices/names; default: from the frame's contrasts
    paired:bool | None=None, # Whether samples are paired; default: from the frame
    impute:bool=True, # Replace zeros with impute_biosynthetic predictions
    min_samples:float=0.2, # Min fraction (0-1) of non-zero samples required
    gamma:float=0.1, # Uncertainty parameter for CLR transform
    custom_scale:float | dict=0, # Ratio of total signal in group2/group1 for an informed scale model (or group_idx: mean(group)/min(mean(groups)) signal dict for multivariate)
    random_state:int | numpy.random._generator.Generator | None=None, # optional random state for reproducibility
    glycan_class:str | None=None, # Glycan class of the glycoforms ('N', 'O', ...); N-glycan type features (complex, high_Man, hybrid, core/antennary fucose, bisecting) only enter the model for 'N'; default: the dataset's glycan_class metadata, else 'N'
)->pandas.DataFrame: # DataFrame with GLM coefficients and FDR-corrected p-values

Analyzes site-specific glycosylation changes in glycoproteomics data using generalized linear models (GLM) with compositional data normalization

df_milk = glycoproteomics_data_loader.human_milk_N_PMID34087070

get_glycoshift_per_site(df_milk, ['Colostrum1', 'Colostrum2', 'Colostrum3'], ['Mature1', 'Mature2', 'Mature3'])
Condition_coefficient Condition_corr_pval Condition_significant Hex_Condition_coefficient Hex_Condition_corr_pval Hex_Condition_significant HexNAc_Condition_coefficient HexNAc_Condition_corr_pval HexNAc_Condition_significant Neu5Ac_Condition_coefficient ... complex_Condition_significant dHex_Condition_coefficient dHex_Condition_corr_pval dHex_Condition_significant high_Man_Condition_coefficient high_Man_Condition_corr_pval high_Man_Condition_significant hybrid_Condition_coefficient hybrid_Condition_corr_pval hybrid_Condition_significant
P10909_374 11.003143 2.205473e-24 True -21.866421 4.932419e-25 True 5.571504 4.209452e-23 True -1.469129 ... True 34.478558 9.634171e-26 True 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P01011_106 37.301835 4.544324e-22 True 8.597721 3.240239e-23 True -28.704113 5.098921e-21 True 9.757607 ... True -35.895702 5.235743e-23 True 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P07602_215 11.872231 9.508737e-21 True -33.685935 4.506236e-20 True 29.210907 7.798520e-20 True 5.466445 ... True 11.872231 1.118675e-20 True 9.139009 8.846142e-20 True 0.000000 1.000000e+00 False
P24821_327 1.372925 8.307229e-17 True 6.864626 7.818568e-17 True 5.491701 7.818568e-17 True -40.233617 ... True 1.372925 7.818568e-17 True 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P06858_70 -0.138017 8.952165e-14 True -0.690084 8.425567e-14 True -0.552067 6.018262e-14 True -13.395384 ... True 12.981334 1.321738e-17 True 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P01871_46 2.183549 8.173888e-13 True 1.241158 9.026027e-07 True -3.494675 1.207887e-15 True 14.412420 ... True 2.183549 6.410892e-13 True 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P10909_103 -0.722082 1.740507e-12 True -3.610409 1.638125e-12 True -2.888327 1.433359e-12 True 21.209039 ... True -0.722082 1.433359e-12 True 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P01833_186 1.306143 3.930302e-11 True -5.392580 2.564386e-11 True 5.224572 3.288096e-11 True -1.459286 ... True 4.071572 1.892816e-14 True 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P02765_156 -0.511649 2.353016e-10 True 2.937741 1.881030e-10 True -2.046595 1.993143e-10 True -6.519282 ... True 5.495985 1.534251e-10 True 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P47710_69 0.289541 1.089307e-09 True -1.503953 2.744323e-09 True 1.158164 9.320270e-10 True 0.929697 ... True 3.241199 1.264030e-09 True 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P25311_109 -0.372386 1.277871e-09 True -1.861931 1.322973e-09 True -1.489545 1.102477e-09 True 0.043921 ... True 9.492600 8.580032e-14 True 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P0C0L5_226 -28.680566 2.277092e-09 True 18.111339 2.315163e-09 True -57.361132 1.978288e-09 True 0.000000 ... False 0.000000 1.000000e+00 False -28.680566 9.644155e-09 True 0.000000 1.000000e+00 False
P10909_291 0.125205 5.623001e-09 True 0.626024 5.292236e-09 True -0.990710 2.731725e-09 True 0.125205 ... True 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False 1.491529 6.027501e-08 True
P08571_151 -27.886409 5.623001e-09 True 18.001733 4.425171e-09 True -21.978514 1.399867e-08 True 14.700316 ... True 0.000000 1.000000e+00 False -42.316188 5.548065e-09 True -5.475819 3.473702e-01 False
P01591_71 42.724023 7.122701e-08 True -5.194748 2.724638e-02 True -8.671393 9.720115e-04 True -8.798990 ... True -3.118598 3.356059e-01 False 0.000000 1.000000e+00 False 18.376130 4.723805e-04 True
Q08380_551 -1.912574 9.352811e-08 True -6.200274 1.649353e-13 True 8.783184 6.018262e-14 True 6.304166 ... True 0.000000 1.000000e+00 False -8.216740 2.065116e-12 True 0.000000 1.000000e+00 False
P0DOX2_246 -13.445023 1.950178e-04 True 0.651510 3.291974e-01 False 5.811469 1.482340e-04 True -5.075537 ... True -18.650503 5.289632e-04 True 0.797895 8.727273e-01 False -1.294999 9.454545e-01 False
P01024_85 -10.979048 1.656752e-03 True 9.408833 1.477550e-03 True -21.958095 1.403366e-03 True 0.000000 ... False 0.000000 1.000000e+00 False -10.979048 7.016832e-03 True 0.000000 1.000000e+00 False
P19652_56 -0.001621 4.035872e-03 True -0.008103 3.608545e-03 True -0.006483 3.436709e-03 True -0.003241 ... True 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P02788_156 -13.375488 5.019683e-03 True 4.504494 2.955551e-02 True -2.204410 3.308678e-01 False 1.768800 ... True 6.453011 1.821802e-03 True 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P24821_1018 1.404241 3.220482e-02 True -1.118785 2.917707e-02 True 2.808482 2.893267e-02 True 0.000000 ... False 0.000000 1.000000e+00 False 1.404241 1.193473e-01 False 0.000000 1.000000e+00 False
P0DOX2_207 -12.811609 5.479322e-02 True -3.186345 8.345975e-02 False 7.898115 5.826111e-05 True 7.413943 ... True -0.176823 9.526511e-01 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P07602_101 0.000944 8.213297e-02 False 0.004718 7.408072e-02 True 0.003774 7.408072e-02 True 0.000944 ... False 0.000944 1.307307e-01 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P01833_135 -0.001365 9.783597e-02 False -0.006825 8.499776e-02 False -0.005460 8.499776e-02 False -0.002730 ... False -0.001365 1.521829e-01 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P02788_497 1.412371 1.557445e-01 False -3.966515 1.763508e-01 False 5.649484 1.357251e-01 False -7.775096 ... False 3.402115 5.637362e-01 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
Q6WN34_114 -0.000546 1.566794e-01 False -0.002732 1.369299e-01 False -0.002185 1.369299e-01 False -0.000546 ... False -0.000546 2.396272e-01 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P01877_327 10.413107 1.772138e-01 False -3.797678 4.334942e-04 True 2.782034 5.193531e-01 False -5.635758 ... False 0.000000 1.000000e+00 False 16.048865 1.200267e-04 True 1.190658 9.454545e-01 False
P00709_90 -0.549612 2.178057e-01 False -1.014216 4.777876e-01 False -2.198447 1.979249e-01 False 0.698590 ... False 19.818654 1.366540e-16 True 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P01833_469 -9.830877 2.211357e-01 False 10.360782 5.060322e-05 True -12.661842 7.102502e-06 True -2.014921 ... True 0.601039 7.272727e-01 False 0.000000 1.000000e+00 False -20.706156 4.723805e-04 True
Q08380_125 -0.002017 2.211357e-01 False -0.010087 2.081277e-01 False -0.008070 2.049123e-01 False -0.002017 ... False -0.002017 3.547631e-01 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
Q13410_55 -8.767867 2.467762e-01 False -1.400615 4.245568e-01 False 7.367252 2.049123e-01 False -9.804702 ... False 10.138048 1.304585e-02 True 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P01033_101 -0.000955 2.560696e-01 False -0.004774 2.507795e-01 False -0.003819 2.507795e-01 False -0.000955 ... False -0.000955 4.267827e-01 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P10909_86 0.002258 2.560696e-01 False 0.011290 2.560696e-01 False 0.009032 2.560696e-01 False 0.002258 ... False 0.002258 4.267827e-01 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P01876_340 7.582295 2.560696e-01 False -1.422441 2.312212e-01 False -0.094454 9.663220e-01 False 2.691064 ... False -5.268188 1.307307e-01 False 0.661014 8.727273e-01 False 7.182366 3.473702e-01 False
P0C0L5_1391 -0.002707 2.704897e-01 False -0.013534 2.704897e-01 False -0.010828 2.704897e-01 False -0.002707 ... False -0.002707 4.327835e-01 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
Q9NZJ4_3757 0.001253 2.962280e-01 False 0.007517 2.962280e-01 False 0.005012 2.962280e-01 False 0.001253 ... False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
O60225_152 0.000330 3.273787e-01 False 0.002969 3.273787e-01 False 0.000660 3.273787e-01 False 0.000000 ... False 0.000000 1.000000e+00 False 0.000330 8.727273e-01 False 0.000000 1.000000e+00 False
P24821_1119 -0.001260 4.282784e-01 False -0.006302 4.282784e-01 False -0.005042 4.282784e-01 False -0.001260 ... False -0.001260 6.344866e-01 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P01833_421 -5.302533 4.721212e-01 False -5.091342 1.312784e-01 False 9.695514 4.516998e-02 True -0.536194 ... False -0.170286 9.522628e-01 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P02788_534 -0.000608 4.872411e-01 False -0.003041 4.872411e-01 False -0.002433 4.872411e-01 False -0.000608 ... False -0.000608 6.620121e-01 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P25311_112 0.001661 4.965091e-01 False 0.008304 4.965091e-01 False 0.006643 4.965091e-01 False 0.001661 ... False 0.001661 6.620121e-01 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P00734_143 -0.000248 5.072288e-01 False -0.001242 5.072288e-01 False -0.000994 5.072288e-01 False -0.000497 ... False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P01833_499 -6.190913 5.200281e-01 False 7.179902 3.608545e-03 True -9.919761 7.915336e-02 False -8.425150 ... False 2.179052 6.620121e-01 False -16.857299 2.315925e-02 True -3.635730 9.454545e-01 False
P07602_332 0.000929 5.419882e-01 False 0.001857 5.419882e-01 False 0.001857 5.419882e-01 False 0.000000 ... False 0.000000 1.000000e+00 False 0.000929 8.727273e-01 False 0.000000 1.000000e+00 False
Q6ZMW3_1352 -0.000986 5.560690e-01 False -0.002959 5.560690e-01 False -0.004931 5.560690e-01 False 0.000000 ... False -0.000986 7.175083e-01 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P02790_453 -0.001216 5.800360e-01 False -0.006081 5.800360e-01 False -0.004865 5.800360e-01 False -0.002432 ... False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P02749_253 0.000824 6.283032e-01 False 0.004118 6.283032e-01 False 0.003295 6.283032e-01 False 0.001647 ... False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
Q08431_238 -0.053430 7.776351e-01 False 0.070673 6.905309e-01 False -0.135220 6.801319e-01 False -0.028359 ... False 0.000000 1.000000e+00 False -0.025071 9.231754e-01 False -0.028359 9.454545e-01 False
P00738_241 0.000473 7.923659e-01 False 0.002363 7.923659e-01 False 0.001891 7.923659e-01 False 0.000945 ... False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P02788_521 0.000183 8.036502e-01 False 0.000913 8.036502e-01 False 0.000730 8.036502e-01 False 0.000365 ... False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P07602_426 -0.000476 8.227525e-01 False -0.002379 8.227525e-01 False -0.000952 8.227525e-01 False 0.000000 ... False 0.000000 1.000000e+00 False -0.000476 8.727273e-01 False 0.000000 1.000000e+00 False
P00739_126 -0.000261 8.298991e-01 False -0.001307 8.298991e-01 False -0.001045 8.298991e-01 False -0.000261 ... False -0.000261 8.298991e-01 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P18505_376 0.000512 8.520762e-01 False 0.002558 8.520762e-01 False 0.002046 8.520762e-01 False 0.001023 ... False 0.001535 8.520762e-01 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
Q08380_69 -0.000080 9.493157e-01 False -0.000482 9.493157e-01 False -0.000321 9.493157e-01 False -0.000080 ... False -0.000080 9.493157e-01 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False
P0DOX2_320 0.518476 9.517452e-01 False 1.187979 5.781108e-01 False -0.767261 8.047623e-01 False -15.669119 ... False 6.908342 1.521829e-01 False 0.000000 1.000000e+00 False 0.000000 1.000000e+00 False

55 rows × 27 columns

#glycoproteomics data are recognized by their protein_site_glycan row labels: glycoforms are compared within their glycosite, motifs are quantified per glycosite, and diversity becomes per-site microheterogeneity
print(get_differential_expression(df_milk, motifs = True).head())
get_biodiversity(df_milk, metrics = ['alpha'])[0].head()
Chosen grouping: by_motif_family
     Glycosite    corr p-val  significant  Effect size
0    P00709_90  1.363759e-25         True    -2.172303
14  P0DOX2_246  1.457395e-25         True     0.263281
11  P07602_215  1.762216e-24         True    -0.347576
8   P01877_327  3.570178e-11         True   -25.357349
15  P0DOX2_320  9.034243e-08         True     1.131434
Metric Group1 mean Group2 mean p-val Effect size corr p-val significant
0 Q08431_238_shannon_diversity 0.686377 1.073391 0.000537 17.610020 0.031154 True
1 Q08431_238_simpson_diversity 0.493256 0.649394 0.001246 10.110418 0.036123 True
2 P00709_90_species_richness 1.000000 4.666667 0.008163 7.778175 0.157826 False
3 Q13410_55_species_richness 2.000000 4.666667 0.024817 3.265986 0.271425 False
4 P0DOX2_207_simpson_diversity 0.725047 0.616102 0.030898 -2.685218 0.271425 False

annotate

extract curated motifs, graph features, and sequence features from glycan sequences

annotate_glycan

def annotate_glycan(
    glycan:str | networkx.classes.digraph.DiGraph, # IUPAC-condensed glycan sequence or NetworkX graph
    motifs:pandas.DataFrame | str | list[str] | None=None, # Motif dataframe (motif_name + motif columns), or motif sequence(s); defaults to motif_list
    termini_list:list=[], # Monosaccharide positions: 'terminal', 'internal', or 'flexible'
    gmotifs:list[networkx.classes.digraph.DiGraph] | None=None, # Precalculated motif graphs for speed
    condense:bool=False, # Remove columns with only zeros
)->pandas.DataFrame: # DataFrame with motif counts for the glycan

Counts occurrences of known motifs in a glycan structure using subgraph isomorphism

annotate_glycan("Neu5Ac(a2-3)Gal(b1-4)[Fuc(a1-3)]GlcNAc(b1-2)Man(a1-3)[Gal(b1-4)GlcNAc(b1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc")
motif_name Terminal_LewisX Internal_LewisX LewisY SialylLewisX SulfoSialylLewisX Terminal_LewisA Internal_LewisA LewisB SialylLewisA SulfoLewisA ... Mucin_elongated_core2 Fucoidan Alginate FG XX Difucosylated_core GalFuc_core DisialylLewisC RM2 DisialylLewisA
Neu5Ac(a2-3)Gal(b1-4)[Fuc(a1-3)]GlcNAc(b1-2)Man(a1-3)[Gal(b1-4)GlcNAc(b1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc 0 1 0 1 0 0 0 0 0 0 ... 0 0 0 0 0 0 0 0 0 0

1 rows × 164 columns

annotate_dataset

def annotate_dataset(
    glycans:str | list[str], # Glycan(s) in IUPAC-condensed nomenclature
    motifs:pandas.DataFrame | None=None, # Motif dataframe (name + sequence); defaults to motif_list
    feature_set:str | list[str]=['known'], # Feature types to analyze: known, graph, exhaustive, terminal(1-3), custom, chemical, size_branch
    termini_list:list=[], # Monosaccharide positions: 'terminal', 'internal', or 'flexible'
    condense:bool=False, # Remove columns with only zeros
    custom_motifs:str | list[str]=[], # Custom motifs (sequences, r-prefixed glyco-regexes, or motif_list names) when using 'custom' feature set; columns keep these labels
)->pandas.DataFrame: # DataFrame mapping glycans to motif counts and other features

Comprehensive glycan annotation combining multiple feature types: structural motifs, graph properties, terminal sequences

glycans = ['Man(a1-3)[Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc',
           'Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-3)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc',
           'GalNAc(a1-4)GlcNAcA(a1-4)[GlcN(b1-7)]Kdo(a2-5)[Kdo(a2-4)]Kdo(a2-6)GlcN4P(b1-6)GlcN4P']
print("Annotate Test")
out = annotate_dataset(glycans)
Annotate Test
motif_name Terminal_LewisX Internal_LewisX LewisY SialylLewisX SulfoSialylLewisX Terminal_LewisA Internal_LewisA LewisB SialylLewisA SulfoLewisA H_antigen_type2 H_antigen_type1 H_antigen_type3 A_antigen A_antigen_type1 A_antigen_type2 A_antigen_type3 B_antigen B_antigen_type1 B_antigen_type2 ExtB Galili_antigen GloboH Gb5 Gb4 Gb3 3SGb3 8DSGb3 3SGb4 8DSGb4 6DSGb4 3SGb5 8DSGb5 6DSGb5 6DSGb5_2 6SGb3 8DSGb3_2 6SGb4 8DSGb4_2 6SGb5 8DSGb5_2 66DSGb5 Forssman_antigen iGb3 I_antigen i_antigen PI_antigen Chitobiose Trimannosylcore Internal_LacNAc_type1 Terminal_LacNAc_type1 Internal_LacNAc_type2 Terminal_LacNAc_type2 Internal_LacdiNAc_type1 Terminal_LacdiNAc_type1 Internal_LacdiNAc_type2 Terminal_LacdiNAc_type2 bisectingGlcNAc VIM PolyLacNAc Ganglio_Series Lacto_Series(LewisC) NeoLacto_Series betaGlucan KeratanSulfate Hyaluronan Mollu_series Arthro_series Cellulose_like Chondroitin_4S GPI_anchor Isoglobo_series LewisD Globo_series Sda SDA Muco_series Heparin Peptidoglycan Dermatansulfate CAD Lactosylceramide Lactotriaosylceramide LexLex GM3 H_type3 GM2 GM1 cisGM1 VIM2 GD3 GD1a GD2 GD1b SDLex Fuc_LN3 GT1b GD1 GD1a_2 LcGg4 GT3 Disialyl_T_antigen GT1a GT2 GT1c 2Fuc_GM1 GQ1c O_linked_mannose GT1aa GQ1b HNK1 GQ1ba O_mannose_Lex 2Fuc_GD1b Sialopentaosylceramide Sulfogangliotetraosylceramide B-GM1 GQ1aa bisSulfo-Lewis x para-Forssman core_fucose core_fucose(a1-3) GP1c B-GD1b GP1ca Isoglobotetraosylceramide polySia high_mannose Gala_series LPS_core Nglycan_complex Oglycan_core1 Oglycan_core2 Oglycan_core3 Oglycan_core4 Oglycan_core5 Oglycan_core6 Oglycan_core7 Xylogalacturonan Sialosylparagloboside LDNF OFuc Arabinogalactan_type2 EGF_repeat Nglycan_hybrid Arabinan Xyloglucan Acharan_Sulfate M3FX M3X 1-6betaGalactan Arabinogalactan_type1 Galactomannan Tetraantennary_Nglycan Mucin_elongated_core2 Fucoidan Alginate FG XX Difucosylated_core GalFuc_core DisialylLewisC RM2 DisialylLewisA
Man(a1-3)[Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-3)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
GalNAc(a1-4)GlcNAcA(a1-4)[GlcN(b1-7)]Kdo(a2-5)[Kdo(a2-4)]Kdo(a2-6)GlcN4P(b1-6)GlcN4P 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

quantify_motifs

def quantify_motifs(
    df:pandas.DataFrame | str | pathlib.Path, # DataFrame, filepath, or shipped dataset name, with samples as columns, abundances as values
    glycans:list[str] | None=None, # List of IUPAC-condensed glycan sequences or compositions, or glycoproteomics glycoforms as protein_site_glycan; auto-detected from the first column, or the index, if None
    feature_set:str | list[str]=['known', 'exhaustive'], # Feature types to analyze: known, graph, exhaustive, terminal(1-3), custom, chemical, size_branch
    custom_motifs:str | list[str]=[], # Custom motifs (sequences, r-prefixed glyco-regexes, or motif_list names) when using 'custom' feature set
    remove_redundant:bool=True, # Remove redundant motifs via deduplicate_motifs
)->pandas.DataFrame: # DataFrame with motif abundances (motifs as rows, samples as columns); for glycoforms, one row per glycosite and motif as protein_site_motif

Extracts and quantifies motif abundances from glycan abundance data by weighting motif occurrences; glycoproteomics glycoforms are quantified per glycosite

quantify_motifs(test_df.iloc[:, 1:], test_df.iloc[:, 0].values.tolist(), ['known', 'exhaustive'])
control_1 tumor_1 control_2 tumor_2 control_3 tumor_3 control_4 tumor_4 control_5 tumor_5 ... control_16 tumor_16 control_17 tumor_17 control_18 tumor_18 control_19 tumor_19 control_20 tumor_20
H_antigen_type2 1.347737 0.892651 2.468405 1.810795 1.589162 0.449339 2.640132 0.572828 2.763890 0.737076 ... 1.070249 0.647786 1.440912 1.810304 1.722289 1.475260 4.847788 4.552496 0.480035 0.494123
Internal_LacNAc_type2 8.845085 10.063160 13.435501 28.834006 5.585973 11.359659 11.672584 21.193308 12.734919 28.597709 ... 10.883437 17.991155 21.166792 16.161351 11.909325 29.924308 12.820872 19.107379 8.802443 10.268911
Terminal_LacNAc_type2 52.982192 13.183951 24.413523 12.870782 9.555884 9.822266 12.628910 13.916662 26.569737 10.733867 ... 18.779972 12.157928 14.828507 20.879287 27.689619 10.734756 28.328965 37.870847 14.835019 8.910804
Disialyl_T_antigen 20.803836 36.895471 32.803297 20.401157 33.971366 30.150599 37.703636 24.728411 31.798990 15.989214 ... 46.337629 39.476930 39.087708 40.348217 35.791797 22.968160 11.026029 2.613718 44.676379 46.125360
Oglycan_core1 37.329013 75.567842 59.998893 57.608119 83.293693 78.436161 73.308916 64.356888 58.197862 60.329536 ... 68.269613 68.762287 62.541874 60.699726 58.713271 58.203265 58.826129 42.904325 74.390026 79.515568
Oglycan_core2 61.827277 23.247111 37.849024 41.704788 15.141858 21.181925 24.301494 35.109970 39.304656 39.331576 ... 29.663409 30.149083 35.995300 37.040638 39.598944 40.659064 41.149838 56.978227 23.637462 19.179715
Gal 163.691481 126.500106 141.895063 147.702533 115.056369 132.721945 122.804259 138.398297 141.412183 167.203077 ... 133.838024 140.218313 142.530133 139.697255 138.848449 154.791018 142.588964 157.426027 122.916027 120.555251
Gal3/6S(b1-3)GalNAc 0.843710 1.185047 2.152084 0.687093 1.564450 0.381914 2.389590 0.533142 2.497482 0.338889 ... 2.066978 1.088630 1.462826 2.259636 1.687785 1.137672 0.024033 0.117449 1.972512 1.304717
GalNAc 100.000000 100.000000 100.000000 100.000000 100.000000 100.000000 100.000000 100.000000 100.000000 100.000000 ... 100.000000 100.000000 100.000000 100.000000 100.000000 100.000000 100.000000 100.000000 100.000000 100.000000
GlcNAc6S(b1-6)GalNAc 2.707913 4.438043 6.198123 6.684838 1.478960 11.921934 0.892356 3.821469 4.605009 28.210391 ... 6.241593 11.157860 7.997660 4.916252 0.937290 15.269626 1.463159 0.565249 1.251077 2.680253
Neu5Ac 80.494155 134.094482 120.708503 125.892731 128.626161 137.543517 132.135127 124.740497 118.279272 134.227059 ... 149.089683 152.360772 145.124475 140.251427 125.331418 121.962226 91.599064 72.000898 142.956534 148.579697
Gal(b1-3)GalNAc 99.156290 98.814953 97.847916 99.312907 98.435550 99.618086 97.610410 99.466858 97.502518 99.661111 ... 97.933022 98.911370 98.537174 97.740364 98.312215 98.862328 99.975967 99.882551 98.027488 98.695283
Neu5Ac(a2-3)Gal 57.345927 94.670033 83.675402 103.574200 91.775344 106.231617 90.136699 98.461821 81.110136 117.087919 ... 97.928245 109.749014 101.760261 93.222423 86.403840 96.715461 80.029183 69.040921 95.565848 99.973512
Neu5Ac(a2-6)GalNAc 23.063482 39.304399 36.644881 22.263129 36.571122 31.229766 41.628644 26.256121 37.088978 17.054227 ... 50.675599 41.982557 42.829042 46.391984 38.682564 25.118814 11.540028 2.937334 47.171520 48.274238
Neu5Ac(a2-8)Neu5Ac 0.084745 0.120050 0.388219 0.055402 0.279696 0.082135 0.369784 0.022555 0.080158 0.084913 ... 0.485839 0.629202 0.535171 0.637019 0.245015 0.127952 0.029853 0.022643 0.219166 0.331947

15 rows × 40 columns

get_motif_dag

def get_motif_dag(
    motifs:list[str], # Motif labels as produced by annotate_dataset/quantify_motifs, optionally per glycosite as protein_site_motif
    abundances:pandas.DataFrame | None=None, # Motifs x samples abundances, used as an exact prefilter for containment (not needed for per-glycosite motifs)
)->networkx.classes.digraph.DiGraph: # Transitively reduced containment DAG; edge parent -> child means parent is a substructure of child

Builds the containment DAG of a motif set, in which a parent motif is a substructure of each of its children; per-glycosite motifs are only ordered within their glycosite

get_composition_dag

def get_composition_dag(
    compositions:list[str], # Composition or IUPAC-condensed labels, optionally prefixed as protein_site_composition
    abundances:pandas.DataFrame | None=None, # Glycoforms x samples abundances, used as an exact prefilter for containment
)->networkx.classes.digraph.DiGraph: # Transitively reduced containment DAG; edge parent -> child means parent is contained in child

Builds the containment DAG of a glycoform set, ordering parts by substructure containment where sequences are given and by component-wise composition dominance otherwise

get_k_saccharides

def get_k_saccharides(
    glycans:str | list[str] | set[str], # Glycan(s) in IUPAC-condensed nomenclature
    size:int=2, # Number of monosaccharides per fragment
    up_to:bool=False, # Include fragments up to size k (adds monosaccharides)
    just_motifs:bool=False, # Return nested list of motifs instead of count DataFrame
    terminal:bool=False, # Only count terminal fragments
)->pandas.DataFrame | list[list[str]]: # DataFrame of k-saccharide counts or list of motifs per glycan

Extracts k-saccharide fragments from glycan sequences with options for different fragment sizes and positions

glycans = ['Man(a1-3)[Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc',
           'Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-3)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc',
           'GalNAc(a1-4)GlcNAcA(a1-4)[GlcN(b1-7)]Kdo(a2-5)[Kdo(a2-4)]Kdo(a2-6)GlcN4P(b1-6)GlcN4P']
out = get_k_saccharides(glycans, size = 3)
  Fuc(a1-3)[GlcNAc(b1-4)]GlcNAc GalNAc(a1-4)GlcNAcA(a1-4)Kdo GlcN(b1-7)Kdo(a2-5)Kdo GlcNAcA(a1-4)Kdo(a2-5)Kdo GlcNAcA(a1-4)[GlcN(b1-7)]Kdo Kdo(a2-4)Kdo(a2-6)GlcN4P Kdo(a2-4)[Kdo(a2-5)]Kdo Kdo(a2-4/5)Kdo(a2-6)GlcN4P Kdo(a2-5)Kdo(a2-6)GlcN4P Kdo(a2-6)GlcN4P(b1-6)GlcN4P Man(a1-2)Man(a1-2)Man Man(a1-2)Man(a1-3)Man Man(a1-2/3)Man(a1-2/3/6)Man Man(a1-3)Man(a1-6)Man Man(a1-3)Man(b1-4)GlcNAc Man(a1-3)[Man(a1-6)]Man Man(a1-3/6)Man(b1-4)GlcNAc Man(a1-6)Man(b1-4)GlcNAc Man(b1-4)GlcNAc(b1-4)GlcNAc Xyl(b1-2)Man(b1-4)GlcNAc Xyl(b1-2)[Man(a1-3)]Man Xyl(b1-2)[Man(a1-3/6)]Man Xyl(b1-2)[Man(a1-6)]Man
0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 2 1 1 1 1 2 1
1 0 0 0 0 0 0 0 0 0 0 1 1 3 1 1 1 2 1 1 0 0 0 0
2 0 1 1 1 1 1 1 2 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0

get_terminal_structures

def get_terminal_structures(
    glycan:str | networkx.classes.digraph.DiGraph, # IUPAC-condensed glycan sequence or NetworkX graph
    size:int=1, # Number of monosaccharides in terminal fragment (1 or higher)
)->list[str]: # List of terminal structures with linkages

Identifies terminal monosaccharide sequences from non-reducing ends of glycan structure

get_terminal_structures("Neu5Ac(a2-3)Gal(b1-4)GlcNAc(b1-2)Man(a1-3)[Neu5Ac(a2-6)Gal(b1-4)GlcNAc(b1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc")
['Neu5Ac(a2-3)', 'Neu5Ac(a2-6)']

get_molecular_properties

def get_molecular_properties(
    glycan_list:str | list[str], # Glycan(s) in IUPAC-condensed nomenclature
    verbose:bool=False, # Print SMILES not found on PubChem
    placeholder:bool=False, # Return dummy values instead of dropping failed requests
    pubchem:bool=False, # Additionally fetch xlogp and complexity, the only two descriptors that cannot be computed from the structure
)->pandas.DataFrame: # DataFrame with molecular parameters

Computes molecular properties of glycans from their own SMILES, optionally enriched with the two empirical descriptors only PubChem has

out = get_molecular_properties(["Neu5Ac(a2-3)Gal(b1-4)GlcNAc(b1-2)Man(a1-3)[Neu5Ac(a2-6)Gal(b1-4)GlcNAc(b1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc"])
  molecular_formula molecular_weight exact_mass monoisotopic_mass tpsa h_bond_donor_count h_bond_acceptor_count rotatable_bond_count charge heavy_atom_count ring_count covalent_unit_count isotope_atom_count defined_atom_stereo_count
Neu5Ac(a2-3)Gal(b1-4)GlcNAc(b1-2)Man(a1-3)[Neu5Ac(a2-6)Gal(b1-4)GlcNAc(b1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc C84H138N6O62 2223.997520 2222.783024 2222.783024 1070.160000 39 62 43 0 152 11 1 0 56

get_glycan_similarity

def get_glycan_similarity(
    glycan1:str | networkx.classes.digraph.DiGraph, # IUPAC-condensed glycan sequence or NetworkX graph
    glycan2:str | networkx.classes.digraph.DiGraph, # IUPAC-condensed glycan sequence or NetworkX graph
    motifs:pandas.DataFrame | None=None, # Motif dataframe (name + sequence); defaults to motif_list
    feature_set:str | list[str]=['known', 'exhaustive', 'terminal'], # Feature types to analyze: known, graph, exhaustive, terminal(1-3), custom, chemical, size_branch
)->float: # Cosine similarity between glycan1 and glycan2

Calculates cosine similarity between two glycans based on their motif count fingerprints

get_glycan_similarity("Neu5Ac(a2-3)Gal(b1-3)[Neu5Ac(a2-6)]GalNAc", "Neu5Ac(a2-3)Gal(b1-4)[Neu5Ac(a2-6)]GlcNAc")
np.float64(0.7276068751089989)

graph

convert glycan sequences to graphs and contains helper functions to search for motifs / check whether two sequences describe the same sequence, etc.

glycan_to_nxGraph

def glycan_to_nxGraph(
    glycan:str, # Glycan in IUPAC-condensed format
    libr:glycowork.glycan_data.loader.HashableDict[str, int] | None=None, # Dictionary of form glycoletter:index
    termini:str='ignore', # How to encode terminal/internal position; options: ignore, calc, provided
    termini_list:tuple[str] | None=None, # List of positions from terminal/internal/flexible
)->networkx.classes.digraph.DiGraph: # NetworkX graph object of glycan; cached and shared between calls, so copy it before modifying (a changed label would change every later result for that glycan)

Wrapper for converting glycans into networkx graphs; also works with floating substituents

print('Glycan to networkx Graph (only edges printed)')
print(glycan_to_nxGraph('Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc').edges())
Glycan to networkx Graph (only edges printed)
[(1, 0), (3, 2), (4, 1), (4, 3), (5, 4), (6, 5), (7, 6), (9, 8), (10, 7), (10, 9)]

graph_to_string

def graph_to_string(
    graph:networkx.classes.digraph.DiGraph, # Glycan graph (assumes root node is the one with the highest index)
    canonicalize:bool=True, # Whether to output canonicalized IUPAC-condensed
    order_by:str='length', # canonicalize by 'length' or 'linkage'
)->str: # IUPAC-condensed glycan string

Convert glycan graph back to IUPAC-condensed format, handling disconnected components

graph_to_string(glycan_to_nxGraph('Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc'))
'Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc'

compare_glycans

def compare_glycans(
    glycan_a:str | networkx.classes.digraph.DiGraph, # First glycan to compare
    glycan_b:str | networkx.classes.digraph.DiGraph, # Second glycan to compare
    return_matches:bool=False, # Whether to return node mapping between glycans
    subsumes:bool=False, # Whether to instead check that glycan_a covers every structure glycan_b can be, i.e., glycan_b is glycan_a or a more specific version of it (Gal(b1-4)GlcNAc for Gal(b1-3/4)GlcNAc, but not vice versa)
)->bool: # True if glycans are same (or glycan_a subsumes glycan_b), False if not

Check whether two glycans are identical, or whether glycan_a subsumes glycan_b

print("Graph Isomorphism Test")
print(compare_glycans('Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc',
                      'Man(a1-6)[Man(a1-3)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc'))
Graph Isomorphism Test
True

subgraph_isomorphism

def subgraph_isomorphism(
    glycan:str | networkx.classes.digraph.DiGraph, # Glycan sequence or graph
    motif:str | networkx.classes.digraph.DiGraph, # Glycan motif sequence or graph
    termini_list:str | list=[], # Monosaccharide position(s) from terminal/internal/flexible
    count:bool=False, # Whether to return count instead of presence/absence
    return_matches:bool=False, # Whether to return matched subgraphs as node lists
)->bool | int | tuple[int, list[list[int]]]: # Boolean presence, count, or (count, matches)

Check if motif exists as subgraph in glycan

print("Subgraph Isomorphism Test")
print(subgraph_isomorphism('Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc',
                           'Fuc(a1-6)GlcNAc'))
Subgraph Isomorphism Test
True

generate_graph_features

def generate_graph_features(
    glycan:str | networkx.classes.digraph.DiGraph, # Glycan sequence or network graph
    glycan_graph:bool=True, # True if input is glycan, False if network
    label:str='network', # Label for output dataframe if glycan_graph=False
)->pandas.DataFrame: # Dataframe of graph features

Compute graph features of glycan or network

generate_graph_features("Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc")
diameter branching nbrLeaves avgDeg varDeg maxDeg nbrDeg4 max_deg_leaves mean_deg_leaves deg_assort ... flow_edgeMax flow_edgeMin flow_edgeAvg flow_edgeVar secorderMax secorderMin secorderAvg secorderVar egap entropyStation
Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc 8 1 3 1.818182 0.330579 3.0 0 1.0 0.272727 -1.850372e-15 ... 0.333333 0.111111 0.217778 0.007289 45.607017 20.736441 31.679285 62.422895 0.037907 -2.376732

1 rows × 49 columns

largest_subgraph

def largest_subgraph(
    glycan_a:str | networkx.classes.digraph.DiGraph, # First glycan
    glycan_b:str | networkx.classes.digraph.DiGraph, # Second glycan
)->str: # Largest common subgraph in IUPAC format

Find the largest connected structure two glycans share, with wildcards and PTMs matching as in compare_glycans

glycan1 = 'Neu5Ac(a2-3)Gal(b1-4)GlcNAc(b1-2)Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc'
glycan2 = 'Man(a1-3)[Man(a1-6)]Man(a1-6)[Man(a1-3)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc'
largest_subgraph(glycan1, glycan2)
'Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc'

ensure_graph

def ensure_graph(
    glycan:str | networkx.classes.digraph.DiGraph, # Glycan in IUPAC-condensed format or as networkx graph
    **kwargs
)->networkx.classes.digraph.DiGraph: # NetworkX graph object of glycan; for a string, the cached graph glycan_to_nxGraph returns

Ensures function compatibility with string glycans and graph glycans

ensure_graph("Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc")
<networkx.classes.digraph.DiGraph>

get_possible_topologies

def get_possible_topologies(
    glycan:str | networkx.classes.digraph.DiGraph, # Glycan with floating substituent
    exhaustive:bool=False, # Whether to allow additions at internal positions
    allowed_disaccharides:set[str] | None=None, # Permitted disaccharides when creating possible glycans
    modification_map:dict[str, set[str]]={'6S': {'GlcNAc', 'Gal'}, '3S': {'Gal'}, '4S': {'GalNAc'}, 'OS': {'GlcNAc', 'GalNAc', 'Gal'}}, # Maps modifications to valid attachments
    return_graphs:bool=False, # Whether to return glycan graphs (otherwise return converted strings)
)->list[str | networkx.classes.digraph.DiGraph]: # List of possible topology strings or graphs

Create possible glycan graphs given a floating substituent

possible_topology_check

def possible_topology_check(
    glycan:str | networkx.classes.digraph.DiGraph, # Glycan with floating substituent
    glycans:list[str | networkx.classes.digraph.DiGraph], # List of glycans to check against
    exhaustive:bool=False, # Whether to allow additions at internal positions
    **kwargs
)->list[str | networkx.classes.digraph.DiGraph]: # List of matching glycans

Check whether glycan with floating substituent could match glycans from a list

possible_topology_check("{Neu5Ac(a2-3)}Gal(b1-4)GlcNAc(b1-6)[Gal(b1-3)]GalNAc",
                       ["Fuc(a1-2)Gal(b1-3)GalNAc", "Neu5Ac(a2-3)Gal(b1-3)[Gal(b1-4)GlcNAc(b1-6)]GalNAc",
                       "Neu5Ac(a2-6)Gal(b1-3)[Gal(b1-4)GlcNAc(b1-6)]GalNAc"])
['Neu5Ac(a2-3)Gal(b1-3)[Gal(b1-4)GlcNAc(b1-6)]GalNAc']

deduplicate_glycans

def deduplicate_glycans(
    glycans:list[str] | set[str], # List/set of glycans to deduplicate
)->list[str]: # Deduplicated list of glycans

Remove duplicate glycans from a list/set, even if they have different strings

deduplicate_glycans(["Fuc(a1-2)Gal(b1-3)GalNAc", "Neu5Ac(a2-3)Gal(b1-4)GlcNAc(b1-6)[Neu5Ac(a2-3)Gal(b1-3)]GalNAc",
                     "Neu5Ac(a2-3)Gal(b1-3)[Neu5Ac(a2-3)Gal(b1-4)GlcNAc(b1-6)]GalNAc", "Neu5Ac(a2-6)Gal(b1-3)[Gal(b1-4)GlcNAc(b1-6)]GalNAc"])
['Fuc(a1-2)Gal(b1-3)GalNAc',
 'Neu5Ac(a2-3)Gal(b1-4)GlcNAc(b1-6)[Neu5Ac(a2-3)Gal(b1-3)]GalNAc',
 'Neu5Ac(a2-6)Gal(b1-3)[Gal(b1-4)GlcNAc(b1-6)]GalNAc']

processing

process IUPAC-condensed glycan sequences into glycoletters etc.

min_process_glycans

def min_process_glycans(
    glycan_list:str | list[str], # Glycan(s) in IUPAC-condensed format
)->list[list[str]]: # List of glycoletter lists

Convert list of glycans into a nested lists of glycoletters

min_process_glycans(['Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc',
                     'Man(a1-2)Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc'])
[['Man', 'a1-3', 'Man', 'a1-6', 'Man', 'b1-4', 'GlcNAc', 'b1-4', 'GlcNAc'],
 ['Man',
  'a1-2',
  'Man',
  'a1-3',
  'Man',
  'a1-6',
  'Man',
  'b1-4',
  'GlcNAc',
  'b1-4',
  'GlcNAc']]

get_lib

def get_lib(
    glycan_list:str | list[str], # Glycan(s) in IUPAC-condensed nomenclature
)->dict[str, int]: # Dictionary of glycoletter:index mappings

Returns dictionary mapping glycoletters to indices

get_lib(['Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc',
                     'Man(a1-2)Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc'])
{'GlcNAc': 0, 'Man': 1, 'a1-2': 2, 'a1-3': 3, 'a1-6': 4, 'b1-4': 5}

expand_lib

def expand_lib(
    libr_in:dict[str, int], # Existing dictionary of glycoletter:index
    glycan_list:str | list[str], # Glycan(s) in IUPAC-condensed nomenclature
)->dict[str, int]: # Updated dictionary with new glycoletters

Updates libr with newly introduced glycoletters

lib1 = get_lib(['Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc',
                     'Man(a1-2)Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc'])
lib2 = expand_lib(lib1, ['Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc'])
lib2
{'GlcNAc': 0, 'Man': 1, 'a1-2': 2, 'a1-3': 3, 'a1-6': 4, 'b1-4': 5, 'Fuc': 6}

presence_to_matrix

def presence_to_matrix(
    df:pandas.DataFrame, # DataFrame with glycan occurrence
    glycan_col_name:str='glycan', # Column name for glycans
    label_col_name:str='Species', # Column name for labels
)->pandas.DataFrame: # Matrix with labels as rows and glycan occurrences as columns

Converts a dataframe with glycan occurrence to absence/presence matrix

out = presence_to_matrix(df_species[df_species.Order == 'Fabales'].reset_index(drop = True),
                         label_col_name = 'Family')
glycan Apif(a1-2)Xyl(b1-2)[Glc6Ac(b1-4)]Glc Ara(a1-2)Ara(a1-6)GlcNAc Ara(a1-2)Glc(b1-2)Ara Ara(a1-2)GlcA Ara(a1-2)[Glc(b1-6)]Glc Ara(a1-3)Gal(b1-6)Gal Ara(a1-6)Glc Araf(a1-3)Araf(a1-5)[Araf(a1-6)Gal(b1-6)Glc(b1-6)Man(a1-3)]Araf(a1-5)Araf(a1-3)Araf(a1-3)Araf Araf(a1-3)Gal(b1-6)Gal D-Apif(b1-2)Glc D-Apif(b1-2)GlcA D-Apif(b1-3)Xyl(b1-2)[Glc6Ac(b1-4)]Glc D-Apif(b1-3)Xyl(b1-4)Rha(a1-2)Ara D-Apif(b1-3)Xyl(b1-4)Rha(a1-2)D-Fuc D-Apif(b1-3)Xyl(b1-4)[Glc(b1-3)]Rha(a1-2)D-Fuc D-Apif(b1-3)[Gal(b1-4)Xyl(b1-4)]Rha(a1-2)D-Fuc D-Apif(b1-3)[Gal(b1-4)Xyl(b1-4)]Rha(a1-2)[Rha(a1-3)]D-Fuc D-Apif(b1-3)[Gal(b1-4)Xyl(b1-4)]Rha(a1-3)D-Fuc D-Apif(b1-6)Glc D-ApifOMe(b1-3)XylOMe(b1-4)RhaOMe(a1-2)D-FucOMe D-ApifOMe(b1-3)XylOMe(b1-4)[GlcOMe(b1-3)]RhaOMe(a1-2)D-FucOMe Fruf(a2-1)[Glc(b1-2)][Glc(b1-3)Glc4Ac6Ac(b1-3)]Glc Fruf(a2-1)[Glc(b1-2)][Glc(b1-3)Glc4Ac6Ac(b1-3)]Glc6Ac Fruf(a2-1)[Glc(b1-2)][Glc(b1-3)Glc6Ac(b1-3)]Glc Fruf(a2-1)[Glc(b1-2)][Glc(b1-3)Glc6Ac(b1-3)]Glc6Ac Fruf(b2-1)Glc3Ac6Ac Fruf(b2-1)Glc4Ac6Ac Fruf(b2-1)Glc6Ac Fruf(b2-1)[Glc(b1-2)]Glc Fruf(b2-1)[Glc(b1-2)][Glc(b1-3)Glc(b1-3)]Glc Fruf(b2-1)[Glc(b1-2)][Glc(b1-3)]Glc6Ac Fruf(b2-1)[Glc(b1-2)][Glc(b1-4)Glc(b1-3)]Glc Fruf(b2-1)[Glc(b1-2)][Glc(b1-4)Glc(b1-3)]Glc6Ac Fruf(b2-1)[Glc(b1-2)][Glc(b1-4)Glc6Ac(b1-3)]Glc Fruf(b2-1)[Glc(b1-2)][Glc(b1-4)Glc6Ac(b1-3)]Glc6Ac Fruf(b2-1)[Glc(b1-2)][Glc6Ac(b1-3)]Glc Fruf(b2-1)[Glc(b1-2)][Glc6Ac(b1-3)]Glc6Ac Fruf(b2-1)[Glc(b1-4)Glc6Ac(b1-3)]Glc6Ac Fruf(b2-1)[Glc3Ac(b1-2)]Glc Fruf(b2-1)[Glc6Ac(b1-2)]Glc Fruf1Ac(b2-1)Glc2Ac4Ac6Ac Fuc(a1-2)Gal(b1-2)Xyl(a1-6)Glc Fuc(a1-2)Gal(b1-2)Xyl(a1-6)Glc(b1-4)Glc Fuc(a1-2)Gal(b1-2)Xyl(a1-6)[Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)]Glc(b1-4)Glc Fuc(a1-2)Gal(b1-2)Xyl(a1-6)[Glc(b1-4)]Glc(b1-4)Glc Fuc(a1-2)Gal(b1-4)Xyl Fuc(a1-3)[Gal(b1-4)]GlcNAc(b1-2)Man(a1-6)[GlcNAc(b1-2)Man(a1-3)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)GlcNAc Fuc(a1-4)GlcNAc(b1-2)Man(a1-3)[Gal(b1-3)[Fuc(a1-4)]GlcNAc(b1-2)Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Fuc(a1-6)GlcNAc(b1-2)[Man(a1-6)]Man(a1-6)[Xyl(b1-2)][Man(a1-3)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(?1-?)Gal(b1-4)[Fuc(a1-3)]GlcNAc(b1-2)Man(a1-3)[Man(a1-2)Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(?1-?)[Gal(?1-?)]GlcNAc(?1-?)[Fuc(a1-3)]GlcNAc(b1-2)Man(a1-3)[Gal(?1-?)Man(a1-3)Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(a1-4)Gal Gal(a1-6)Gal Gal(a1-6)Gal(a1-6)Gal Gal(a1-6)Gal(a1-6)Gal(a1-6)Gal(a1-6)Glc(a1-2)Fru Gal(a1-6)Gal(a1-6)Gal(a1-6)Gal(a1-6)Glc(a1-2)Fruf Gal(a1-6)Gal(a1-6)Gal(a1-6)Gal(a1-6)[Fruf(b2-1)]Glc Gal(a1-6)Gal(a1-6)Gal(a1-6)Glc Gal(a1-6)Gal(a1-6)Gal(a1-6)Glc(a1-2)Fru Gal(a1-6)Gal(a1-6)Gal(a1-6)Glc(a1-2)Fruf Gal(a1-6)Gal(a1-6)Glc Gal(a1-6)Gal(a1-6)Glc(a1-2)Fru Gal(a1-6)Gal(a1-6)Glc(a1-2)Fruf Gal(a1-6)Glc(a1-2)Fru Gal(a1-6)Glc(a1-2)Fruf Gal(a1-6)Man Gal(a1-6)Man(b1-4)Man Gal(a1-6)Man(b1-4)Man(b1-4)Man(b1-4)Man Gal(a1-6)Man(b1-4)Man(b1-4)Man(b1-4)[Gal(a1-6)]Man(b1-4)Man(b1-4)Man(b1-4)[Gal(a1-6)]Man Gal(a1-6)Man(b1-4)Man(b1-4)[Gal(a1-6)]Man Gal(a1-6)Man(b1-4)[Gal(a1-6)]Man Gal(b1-2)Glc Gal(b1-2)GlcA Gal(b1-2)GlcA6Me Gal(b1-2)Xyl(a1-6)Glc(b1-4)[Fuc(a1-2)Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc Gal(b1-2)Xyl(a1-6)[Glc(b1-4)]Glc(b1-4)[Fuc(a1-2)Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc Gal(b1-2)Xyl(a1-6)[Glc(b1-4)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc Gal(b1-2)[Xyl(b1-3)]GlcA Gal(b1-3)GlcNAc(b1-2)Man(a1-3)[Gal(b1-3)GlcNAc(b1-2)Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)GlcNAc(b1-2)Man(a1-3)[Gal(b1-3)[Fuc(a1-4)]GlcNAc(b1-2)Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)GlcNAc(b1-2)Man(a1-3)[GlcNAc(b1-2)Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)GlcNAc(b1-2)Man(a1-3)[Xyl(b1-2)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)GlcNAc(b1-2)Man(a1-6)[GlcNAc(b1-2)Man(a1-3)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)GlcNAc(b1-2)Man(a1-6)[Xyl(b1-2)][Man(a1-3)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)GlcNAc(b1-4)Man(a1-3)[Gal(b1-3)GlcNAc(b1-4)Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)GlcNAc(b1-4)Man(a1-3)[GlcNAc(b1-4)Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)GlcNAc(b1-4)Man(a1-3)[Xyl(b1-2)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)GlcNAc(b1-4)Man(a1-6)[GlcNAc(b1-4)Man(a1-3)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)GlcNAc(b1-4)Man(a1-6)[Xyl(b1-2)][Man(a1-3)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)[Fuc(a1-4)]GlcNAc(b1-2)Man(a1-3)[Gal(b1-3)[Fuc(a1-4)]GlcNAc(b1-2)Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)[Fuc(a1-4)]GlcNAc(b1-2)Man(a1-3)[GlcNAc(b1-2)Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)[Fuc(a1-4)]GlcNAc(b1-2)Man(a1-3)[Xyl(b1-2)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)[Fuc(a1-4)]GlcNAc(b1-2)Man(a1-3/6)[Gal(b1-3)GlcNAc(b1-2)Man(a1-3/6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)[Fuc(a1-4)]GlcNAc(b1-2)Man(a1-3/6)[GlcNAc(b1-2)Man(a1-3/6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)[Fuc(a1-4)]GlcNAc(b1-2)Man(a1-3/6)[Xyl(b1-2)][Man(a1-3/6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)[Fuc(a1-4)]GlcNAc(b1-2)Man(a1-6)[GlcNAc(b1-2)Man(a1-3)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)[Fuc(a1-4)]GlcNAc(b1-2)Man(a1-6)[Xyl(b1-2)][Man(a1-3)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)[Fuc(a1-4)]GlcNAc(b1-2)[Man(a1-6)]Man(a1-6)[Xyl(b1-2)][Man(a1-3)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-3)[Fuc(a1-6)]GlcNAc(b1-2)[Man(a1-6)]Man(a1-6)[Xyl(b1-2)][Man(a1-3)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Gal(b1-4)Gal(b1-4)Man Gal(b1-4)Gal(b1-4)ManOMe Gal(b1-4)GlcA Gal(b1-4)GlcNAc(b1-2)Man(a1-3)[Gal(b1-4)GlcNAc(b1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc Gal(b1-4)GlcNAc(b1-2)[Gal(b1-4)GlcNAc(b1-4)]Man(a1-3)[Gal(b1-4)GlcNAc(b1-2)[Gal(b1-4)GlcNAc(b1-6)]Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Gal(b1-4)Man(b1-4)Man Gal(b1-4)Man(b1-4)Man(b1-4)Gal Gal(b1-4)Xyl(b1-4)Rha(a1-2)D-Fuc Gal(b1-4)Xyl(b1-4)Rha(a1-2)D-Fuc1CoumOMe Gal(b1-4)Xyl(b1-4)Rha(a1-2)D-Fuc1FerOMe Gal(b1-4)Xyl(b1-4)Rha(a1-2)Fuc Gal(b1-4)Xyl(b1-4)Rha(a1-2)Fuc4Ac Gal(b1-4)Xyl(b1-4)Rha(a1-2)[Rha(a1-3)]D-Fuc Gal(b1-4)Xyl(b1-4)Rha(a1-2)[Rha(a1-3)]D-Fuc1CoumOMe Gal(b1-4)Xyl(b1-4)Rha(a1-2)[Rha(a1-3)]D-FucOMeOSin Gal(b1-4)Xyl(b1-4)Rha(a1-2)[Rha(a1-3)]Fuc Gal(b1-4)Xyl(b1-4)[D-Apif(b1-3)]Rha(a1-2)D-Fuc Gal(b1-4)Xyl(b1-4)[D-Apif(b1-3)]Rha(a1-2)D-Fuc1CoumOMe Gal(b1-4)Xyl(b1-4)[D-Apif(b1-3)]Rha(a1-2)[Rha(a1-3)]D-Fuc Gal(b1-4)Xyl(b1-4)[D-Apif(b1-3)]Rha(a1-2)[Rha(a1-3)]D-Fuc1CoumOMe GalA(a1-2)[Araf(a1-5)Araf(a1-4)]Rha(b1-4)GalA GalA(a1-4)GalA(a1-4)GalA GalA(a1-4)GalA(a1-4)GalA(a1-4)GalA(a1-2)Rha(a1-4)GalA(a1-2)Rha(a1-4)GalA(a1-2)GalA GalA(a1-4)GalA(a1-4)GalA(a1-4)GalA(a1-4)GalA(a1-4)GalA(a1-4)GalA(a1-4)GalA(a1-4)GalA(a1-4)GalA(a1-4)GalA(a1-4)GalA GalOMe(b1-2)[XylOMe(b1-3)]GlcAOMe GalOMe(b1-4)XylOMe(b1-4)RhaOMe(a1-2)D-FucOMe GalOMe(b1-4)XylOMe(b1-4)RhaOMe(a1-2)[RhaOMe(a1-3)]D-FucOMe GalOMe(b1-4)XylOMe(b1-4)[D-ApifOMe(b1-3)]RhaOMe(a1-2)[RhaOMe(a1-3)]D-FucOMe Galf(b1-2)[Galf(b1-4)]Man Glc(a1-2)Fru Glc(a1-2)Glc(a1-3)Glc(a1-3)Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-3)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Glc(a1-2)Glc(a1-3)Glc(a1-3)Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-3)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Glc(a1-2)Glc(a1-3)Glc(a1-3)Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Glc(a1-2)Glc(a1-3)Glc(a1-3)Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAcN Glc(a1-2)Rha(a1-6)Glc Glc(a1-3)Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-3)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Glc(a1-3)Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-6)]Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Glc(a1-3)Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-3)[Man(a1-6)]Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Glc(a1-4)Glc(a1-2)Rha(a1-6)Glc Glc(a1-4)Glc(a1-4)Glc(a1-6)Glc Glc(a1-4)Glc(a1-4)GlcA Glc(a1-4)GlcA(b1-2)GlcA Glc(b1-2)Ara Glc(b1-2)Ara(a1-2)GlcA Glc(b1-2)Gal(b1-2)Gal(b1-2)GlcA Glc(b1-2)Gal(b1-2)GlcA Glc(b1-2)Gal(b1-2)GlcA(b1-3)[Glc(b1-3)]Ara Glc(b1-2)Glc Glc(b1-2)Glc(a1-2)Fru Glc(b1-2)Glc(a1-2)FrufOBzOCin Glc(b1-2)Glc(b1-2)Glc Glc(b1-2)GlcA Glc(b1-2)Xyl Glc(b1-2)[Ara(a1-3)]GlcA6Me Glc(b1-2)[Ara(a1-3)]GlcAOMe Glc(b1-2)[Ara(a1-6)]Glc Glc(b1-2)[Glc(b1-3)]Glc(a1-2)Fruf Glc(b1-2)[Glc(b1-3)]Glc1Fer6Ac(a1-2)Fruf1FerOBz Glc(b1-2)[Glc(b1-3)]Glc6Ac(a1-2)Fru Glc(b1-2)[Glc6Ac(b1-3)]Glc(a1-2)Fru Glc(b1-2)[Glc6Ac(b1-3)]Glc1Fer(a1-2)Fruf1FerOBz Glc(b1-2)[Glc6Ac(b1-3)]Glc6Ac(a1-2)Fru Glc(b1-2)[Rha(a1-3)]GlcA Glc(b1-2)[Xyl(b1-2)Ara(a1-6)]Glc Glc(b1-2)[Xyl(b1-2)D-Fuc(b1-6)]Glc Glc(b1-3)Ara Glc(b1-3)Glc Glc(b1-3)Glc(b1-3)[Glc(b1-2)]Glc(a1-2)Fru Glc(b1-3)Glc(b1-3)[Glc(b1-2)]Glc(a1-2)Fruf Glc(b1-3)Glc6Ac(b1-3)[Glc(b1-2)]Glc(a1-2)Fru Glc(b1-3)Glc6Ac(b1-3)[Glc(b1-2)]Glc(a1-2)Fruf Glc(b1-3)Glc6Ac(b1-3)[Glc(b1-2)]Glc1Coum6Ac(a1-2)Fruf1CoumOBz Glc(b1-3)Glc6Ac(b1-3)[Glc(b1-2)]Glc1Fer(a1-2)Fruf1CoumOBz Glc(b1-3)Glc6Ac(b1-3)[Glc(b1-2)]Glc1Fer(a1-2)Fruf1FerOBz Glc(b1-3)Glc6Ac(b1-3)[Glc(b1-2)]Glc1Fer6Ac(a1-2)Fruf1CoumOBz Glc(b1-3)Glc6Ac(b1-3)[Glc(b1-2)]Glc1Fer6Ac(a1-2)Fruf1FerOBz Glc(b1-3)Glc6Ac(b1-3)[Glc(b1-2)]Glc6Ac(a1-2)Fru Glc(b1-3)Glc6Ac(b1-3)[Glc(b1-2)][Rha(a1-4)]Glc1Coum6Ac(a1-2)Fruf1CoumOBz Glc(b1-3)Glc6Ac(b1-3)[Glc(b1-2)][Rha(a1-4)]Glc1Fer6Ac(a1-2)Fruf1CoumOBz Glc(b1-3)Rha1Fer(a1-4)Fruf(b2-1)GlcOBz Glc(b1-3)[Araf(a1-4)]Rha(a1-2)Glc Glc(b1-3)[Xyl(b1-4)]Rha(a1-2)D-FucOMe Glc(b1-4)Glc(b1-3)[Glc(b1-2)]Glc(a1-2)Fru Glc(b1-4)Glc(b1-3)[Glc(b1-2)]Glc(a1-2)Fruf Glc(b1-4)Glc(b1-3)[Glc(b1-2)]Glc1Coum6Ac(a1-2)Fruf1FerOBz Glc(b1-4)Glc(b1-3)[Glc(b1-2)]Glc1Fer(a1-2)Fruf1FerOBz Glc(b1-4)Glc(b1-3)[Glc(b1-2)]Glc1Fer6Ac(a1-2)Fruf1CoumOBz Glc(b1-4)Glc(b1-3)[Glc(b1-2)]Glc1Fer6Ac(a1-2)Fruf1FerOBz Glc(b1-4)Glc(b1-3)[Glc(b1-2)]Glc6Ac(a1-2)Fru Glc(b1-4)Glc(b1-4)Glc Glc(b1-4)Glc(b1-4)Glc(b1-4)Man Glc(b1-4)Glc6Ac(b1-3)Glc1Fer6Ac(a1-2)Fruf1FerOBz Glc(b1-4)Glc6Ac(b1-3)Glc6Ac(a1-2)Fru Glc(b1-4)Glc6Ac(b1-3)[Glc(b1-2)]Glc(a1-2)Fru Glc(b1-4)Glc6Ac(b1-3)[Glc(b1-2)]Glc1Coum6Ac(a1-2)Fruf1FerOBz Glc(b1-4)Glc6Ac(b1-3)[Glc(b1-2)]Glc1Fer(a1-2)Fruf1FerOBz Glc(b1-4)Glc6Ac(b1-3)[Glc(b1-2)]Glc1Fer6Ac(a1-2)Fruf1CoumOBz Glc(b1-4)Glc6Ac(b1-3)[Glc(b1-2)]Glc1Fer6Ac(a1-2)Fruf1FerOBz Glc(b1-4)Glc6Ac(b1-3)[Glc(b1-2)]Glc6Ac(a1-2)Fru Glc(b1-4)Man(b1-4)Glc Glc(b1-4)Rha Glc(b1-4)Rha1Fer(a1-4)Fruf(b2-1)GlcOBz Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc Glc(b1-6)Glc(b1-3)Glc Glc(b1-6)GlcNAc Glc1Cer Glc2Ac(b1-4)[D-Apif(b1-3)Xyl(b1-2)]Glc Glc2Ac3Ac4Ac6Ac(b1-3)Ara Glc3Ac(b1-2)Glc(a1-2)Fru Glc6Ac(a1-2)Fru Glc6Ac(b1-2)Glc(a1-2)Fru Glc6Ac(b1-2)Glc(a1-2)FrufOBzOCin Glc6Ac(b1-3)Ara Glc6Ac(b1-3)Glc6Ac(b1-3)[Glc6Ac(b1-2)]Glc1Fer6Ac(a1-2)Fruf1CoumOAcOBz Glc6Ac(b1-3)Glc6Ac(b1-3)[Glc6Ac(b1-2)][RhaOAc(a1-4)]Glc1Fer6Ac(a1-2)Fruf1CoumOAcOBz Glc6Ac(b1-3)[Glc(b1-2)]Glc1Coum(a1-2)Fruf1CoumOBz Glc6Ac(b1-3)[Glc(b1-2)]Glc1Fer(a1-2)Fruf1CoumOBz Glc6Ac(b1-3)[Glc(b1-2)]Glc1Fer(a1-2)Fruf1FerOBz Glc6Ac(b1-3)[Glc(b1-2)]Glc1Fer6Ac(a1-2)Fruf1FerOBz GlcA(b1-2)Glc GlcA(b1-2)GlcA GlcA(b1-2)GlcA(b1-2)Rha GlcA4Me(a1-2)[Xyl(b1-4)Xyl(b1-4)Xyl(b1-4)Xyl(b1-4)]Xyl GlcA4Me(a1-2)[Xyl(b1-4)Xyl(b1-4)Xyl(b1-4)]Xyl GlcA4Me(a1-2)[Xyl(b1-4)]Xyl GlcNAc(b1-2)Man(a1-3)[Gal(b1-3)GlcNAc(b1-2)Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc GlcNAc(b1-2)Man(a1-3)[GlcNAc(b1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc GlcNAc(b1-2)Man(a1-3)[GlcNAc(b1-2)Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)GlcNAc GlcNAc(b1-2)Man(a1-3)[GlcNAc(b1-2)Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc GlcNAc(b1-2)Man(a1-3)[GlcNAc(b1-2)Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Gal(a1-3)]GlcNAc GlcNAc(b1-2)Man(a1-3)[Man(a1-3)[Man(a1-6)]Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)GlcNAc GlcNAc(b1-2)Man(a1-3)[Man(a1-3)[Man(a1-6)]Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc GlcNAc(b1-2)Man(a1-3)[Xyl(b1-2)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc GlcNAc(b1-2)Man(a1-3)[Xyl(b1-2)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc GlcNAc(b1-2)Man(a1-3/6)[Xyl(b1-2)][Man(a1-3/6)]Man(b1-4)GlcNAc(b1-4)GlcNAc GlcNAc(b1-2)Man(a1-3/6)[Xyl(b1-2)][Man(a1-3/6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc GlcNAc(b1-2)Man(a1-6)[Man(a1-3)]Man(b1-4)GlcNAc(b1-4)GlcNAc GlcNAc(b1-2)Man(a1-6)[Man(a1-3)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc GlcNAc(b1-2)Man(a1-6)[Xyl(b1-2)][Man(a1-3)]Man(b1-4)GlcNAc(b1-4)GlcNAc GlcNAc(b1-2)Man(a1-6)[Xyl(b1-2)][Man(a1-3)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc GlcNAc(b1-4)Man(a1-3)[GlcNAc(b1-4)Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc GlcNAc(b1-4)Man(a1-3)[Xyl(b1-2)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc GlcNAc(b1-4)Man(a1-3)[Xyl(b1-2)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc GlcNAc(b1-4)Man(a1-6)[Xyl(b1-2)][Man(a1-3)]Man(b1-4)GlcNAc(b1-4)GlcNAc GlcNAc(b1-4)Man(a1-6)[Xyl(b1-2)][Man(a1-3)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc GlcOMe(b1-3)[XylOMe(b1-4)]RhaOMe(a1-2)D-FucOMe Glcf(b1-2)Xyl(b1-4)Rha(b1-4)[Xyl(b1-3)]Xyl Hexf(?1-?)Xyl(b1-4)Rha(b1-4)[Xyl(a1-3)]Xyl L-Lyx(a1-2)Ara(a1-2)GlcA Lyx(a1-2)Ara(a1-2)GlcA Man(a1-2)Man(a1-2)Man(a1-2)Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-2)Man(a1-2)Man(a1-2)Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAcN Man(a1-2)Man(a1-2)Man(a1-2)Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-2)Man(a1-2)Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-6)]Man(a1-6)]Man(b1-4)GlcNAc Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-6)]Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-3)[Man(a1-6)]Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-6)[Man(a1-3)]Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-3)[Man(a1-6)]Man(a1-6)]Man(b1-4)GlcNAc Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-3)[Man(a1-6)]Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-3)[Man(a1-6)]Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAcN Man(a1-2)Man(a1-2)Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-2)Man(a1-2)Man(a1-6)[Man(a1-2)Man(a1-3)]Man(a1-3)[Man(a1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-2)Man(a1-2)[Man(a1-6)]Man(a1-3)[Man(a1-2)Man(a1-6)[Man(a1-3)]Man(a1-6)]Man(b1-4)GlcNAc Man(a1-2)Man(a1-3)Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-2)Man(a1-6)]Man(a1-3)[Man(a1-2)Man(a1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-6)]Man(a1-3)[Man(a1-2)Man(a1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-2)Man(a1-3)[Man(a1-2)Man(a1-6)]Man(a1-6)[Man(a1-2)Man(a1-3)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-2)Man(a1-3)[Man(a1-3)[Man(a1-6)]Man(a1-6)]Man(b1-4)GlcNAc Man(a1-2)Man(a1-3)[Man(a1-3)[Man(a1-6)]Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-2)Man(a1-3)[Man(a1-6)]Man(a1-6)[Man(a1-2)Man(a1-2)Man(a1-3)]Man(b1-4)GlcNAc(b1-4)GlcNAcN Man(a1-2)Man(a1-3)[Man(a1-6)]Man(a1-6)[Man(a1-2)Man(a1-3)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-2)Man(a1-3)[Man(a1-6)]Man(a1-6)[Man(a1-2)Man(a1-3)]Man(b1-4)GlcNAc(b1-4)GlcNAcN Man(a1-2)Man(a1-6)[Man(a1-2)Man(a1-3)]Man(a1-6)[Man(a1-2)Man(a1-2)Man(a1-3)]Man(b1-4)GlcNAc(b1-4)GlcNAcN Man(a1-2)Man(a1-6)[Man(a1-3)]Man(a1-3)[Man(a1-2)Man(a1-6)[Man(a1-3)]Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAcN Man(a1-2)Man(a1-6)[Man(a1-3)]Man(a1-6)[Man(a1-2)Man(a1-3)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-2)Man(a1-6)[Man(a1-3)]Man(a1-6)[Man(a1-3)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-2)[Man(a1-3)]Man(a1-6)[Xyl(b1-2)][Man(a1-3)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Man(a1-3)Man(a1-6)[Man(a1-3)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-3)Man(a1-6)[Xyl(b1-2)][Man(a1-3)]Man(a1-6)Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-3)Man(a1-6)[Xyl(b1-2)][Man(a1-3)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-3)Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-3)Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Man(a1-3)[Man(a1-2)Man(a1-6)]Man(a1-3)[Man(a1-3)[Man(a1-2)Man(a1-6)]Man(a1-6)][Xyl(b1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Man(a1-3)[Man(a1-2)Man(a1-6)]Man(a1-6)[Man(a1-2)Man(a1-2)Man(a1-3)]Man(b1-4)GlcNAc(b1-4)GlcNAcN Man(a1-3)[Man(a1-6)]Man(a1-6)[Man(a1-3)]Man(b1-4)GlcNAc Man(a1-3)[Man(a1-6)]Man(a1-6)[Man(a1-3)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-3)[Man(a1-6)]Man(a1-6)[Xyl(b1-2)][Man(a1-3)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Man(a1-3)[Man(a1-6)][Xylf(a1-2)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Man(a1-3)[Xyl(b1-2)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc-ol Man(a1-3)[Xyl(b1-2)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAcN Man(a1-3)[Xyl(b1-2)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]Hex Man(a1-3)[Xylf(b1-2)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Man(a1-3/6)Man(a1-6)[Man(a1-3)]Man(b1-4)GlcNAc Man(a1-3/6)Man(a1-6)[Man(a1-3)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-3/6)Man(a1-6)[Xyl(b1-2)][Man(a1-3)]Man(b1-4)GlcNAc(b1-4)GlcNAc Man(a1-3/6)Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Man(b1-2)Man Man(b1-4)Gal(b1-4)Gal(b1-4)Man Man(b1-4)Gal(b1-4)Gal(b1-4)ManOMe Man(b1-4)Man Man(b1-4)Man(b1-4)Man Man(b1-4)Man(b1-4)Man(b1-4)Man Man(b1-4)Man(b1-4)Man(b1-4)Man(b1-4)Man Man(b1-4)Man(b1-4)Man(b1-4)[Gal(a1-6)]Man Man(b1-4)Man(b1-4)[Gal(a1-6)]Man Man(b1-4)Man(b1-4)[Gal(a1-6)]Man(b1-4)[Gal(a1-6)]Man(b1-4)Man(b1-4)Man(b1-4)Man Man(b1-4)Man(b1-4)[Gal(a1-6)]Man(b1-4)[Gal(a1-6)]Man(b1-4)Man(b1-4)Man(b1-4)[Man(b1-6)]Man(b1-4)[Man(b1-6)]Man(b1-4)Man(b1-4)Man(b1-4)[Gal(a1-6)]Man(b1-4)[Gal(a1-3)Gal(a1-3)Gal(a1-6)]Man(b1-4)Man(b1-4)Man(b1-4)[Man(b1-6)]Man(b1-4)[Man(b1-6)]Man(b1-4)Man(b1-4)Man(b1-4)[Man(b1-6)]Man(b1-4)[Man(b1-6)]Man(b1-4)Man(b1-4)Man Man(b1-4)[Gal(a1-6)]Man Man(b1-4)[Gal(a1-6)]Man(b1-4)Man Man(b1-4)[Gal(a1-6)]Man(b1-4)Man(b1-4)Man Man(b1-4)[Gal(a1-6)]Man(b1-4)[Gal(a1-6)]Man Man(b1-4)[Gal(a1-6)]Man(b1-4)[Gal(a1-6)]Man(b1-4)[Gal(a1-6)]Man(b1-4)Man Man(b1-4)[Gal(a1-6)]Man(b1-4)[Gal(a1-6)]Man(b1-4)[Gal(a1-6)]Man(b1-4)[Gal(a1-6)]Man Man(b1-6)Glc Neu5Ac(a2-6)Gal(b1-4)GlcNAc(b1-2)Man(a1-3)[Neu5Ac(a2-6)Gal(b1-4)GlcNAc(b1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc Neu5Ac(a2-6)Gal(b1-4)GlcNAc(b1-2)[Neu5Ac(a2-3)Gal(b1-4)GlcNAc(b1-4)]Man(a1-3)[Neu5Ac(a2-6)Gal(b1-4)GlcNAc(b1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Rha(a1-2)Ara Rha(a1-2)Ara(a1-2)GlcA Rha(a1-2)Ara(a1-2)GlcA6Me Rha(a1-2)Ara(a1-2)GlcAOMe Rha(a1-2)D-Ara(b1-2)GlcA Rha(a1-2)Gal(b1-2)Glc Rha(a1-2)Gal(b1-2)GlcA Rha(a1-2)Gal(b1-2)GlcA6Me Rha(a1-2)Gal(b1-2)GlcAOMe Rha(a1-2)Glc Rha(a1-2)Glc(b1-2)Glc Rha(a1-2)Glc(b1-2)GlcA Rha(a1-2)Glc(b1-2)GlcA6Me Rha(a1-2)Glc(b1-2)GlcAOMe Rha(a1-2)Glc(b1-6)Glc Rha(a1-2)GlcA(b1-2)GlcA Rha(a1-2)GlcAOMe(b1-2)GlcAOMe Rha(a1-2)Rha(a1-2)Gal(b1-4)[Glc(b1-2)]GlcA Rha(a1-2)Xyl Rha(a1-2)Xyl(b1-2)Glc Rha(a1-2)Xyl(b1-2)GlcA Rha(a1-2)Xyl(b1-2)GlcA6Me Rha(a1-2)Xyl(b1-2)GlcAOMe Rha(a1-2)Xyl3Ac Rha(a1-2)Xyl4Ac Rha(a1-2)[Glc(b1-3)]Glc Rha(a1-2)[Glc(b1-6)]Gal(b1-2)GlcA6Me Rha(a1-2)[Rha(a1-4)]Glc Rha(a1-2)[Rha(a1-6)]Gal Rha(a1-2)[Rha(a1-6)]Glc Rha(a1-2)[Xyl(b1-4)]Glc Rha(a1-2)[Xyl(b1-4)]Glc(b1-6)Glc Rha(a1-3)GlcA Rha(a1-3)[Rha(a1-4)]Gal Rha(a1-4)Gal(b1-2)GlcA Rha(a1-4)Gal(b1-2)GlcAOMe Rha(a1-4)Gal(b1-2)GlcOMe Rha(a1-4)Gal(b1-4)Gal(b1-4)GalGro Rha(a1-4)Xyl(b1-2)Glc Rha(a1-4)Xyl(b1-2)GlcA Rha(a1-4)Xyl(b1-2)GlcAOMe Rha(a1-6)Glc Rha(a1-6)[Xyl(b1-3)Xyl(b1-2)]Glc(b1-2)Glc Rha(b1-2)Glc(b1-2)GlcA Rha1Fer(a1-4)Fruf(b2-1)GlcOBz RhaOMe(a1-2)[RhaOMe(a1-6)]GlcOMe-ol RhaOMe(a1-6)GlcOMe(b1-2)GlcOMe-ol Xyl(a1-2)[Man(a1-3)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Xyl(a1-3)Fuc(a1-4)Rha Xyl(a1-6)Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)Glc-ol Xyl(a1-6)Glc(b1-4)[Fuc(a1-2)Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc-ol Xyl(a1-6)Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Fuc(a1-2)Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc-ol Xyl(a1-6)Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc-ol Xyl(a1-6)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc-ol Xyl(a1-6)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Fuc(a1-2)Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc-ol Xyl(a1-6)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc Xyl(a1-6)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc-ol Xyl(b1-2)Ara(a1-6)Glc Xyl(b1-2)Ara(a1-6)GlcNAc Xyl(b1-2)Ara(a1-6)[Glc(b1-2)]Glc Xyl(b1-2)Ara(a1-6)[Glc(b1-4)]GlcNAc Xyl(b1-2)D-Fuc(b1-6)Glc Xyl(b1-2)D-Fuc(b1-6)GlcNAc Xyl(b1-2)D-Fuc(b1-6)[Glc(b1-2)]Glc Xyl(b1-2)Fuc(a1-6)Glc Xyl(b1-2)Fuc(a1-6)GlcNAc Xyl(b1-2)Fuc(b1-6)Glc Xyl(b1-2)Fuc(b1-6)GlcNAc Xyl(b1-2)Fuc(b1-6)[Glc(b1-2)]Glc Xyl(b1-2)Gal(b1-2)GlcA6Me Xyl(b1-2)Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Xyl(b1-2)Rha(a1-2)Ara Xyl(b1-2)Xyl(b1-3)[Rha(b1-2)Rha(b1-4)]Xyl Xyl(b1-2)[Glc(b1-3)]Ara Xyl(b1-2)[Glc2Ac(b1-4)]Glc Xyl(b1-2)[Man(a1-3)][Man(a1-6)]Man(a1-3)Man(b1-4)GlcNAc(b1-4)GlcNAc Xyl(b1-2)[Man(a1-3)][Man(a1-6)]Man(a1-3)Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Xyl(b1-2)[Man(a1-3)][Man(a1-6)]Man(a1-6)Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Xyl(b1-2)[Man(a1-3)][Man(a1-6)]Man(b1-4)GlcNAc Xyl(b1-2)[Man(a1-3)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Xyl(b1-2)[Man(a1-3)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Xyl(b1-2)[Man(a1-3)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc-ol Xyl(b1-2)[Man(a1-3)][Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc Xyl(b1-2)[Man(a1-3)][Man(a1-6)]Man(b1-4)ManNAc Xyl(b1-2)[Man(a1-6)]Man(a1-3)Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Xyl(b1-2)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc Xyl(b1-2)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-3)]GlcNAc Xyl(b1-2)[Rha(a1-3)]GlcA Xyl(b1-3)Ara Xyl(b1-3)Xyl(b1-2)[Rha(a1-6)]Glc(b1-2)Glc Xyl(b1-3)Xyl(b1-4)Rha(a1-2)[Rha(a1-6)]Glc Xyl(b1-3)Xyl(b1-4)Rha(a1-2)[Rha(a1-6)]Glc(b1-2)Glc Xyl(b1-4)Rha(a1-2)Ara Xyl(b1-4)Rha(a1-2)D-Fuc Xyl(b1-4)Rha(a1-2)D-FucOMe Xyl(b1-4)Rha(a1-2)Fuc Xyl(b1-4)Rha(a1-2)Fuc3Ac Xyl(b1-4)Rha(a1-2)Fuc4Ac Xyl(b1-4)Rha(a1-2)Glc Xyl(b1-4)Rha(a1-2)[Rha(a1-3)]Fuc4Ac Xyl(b1-4)Rha(a1-2)[Rha(a1-6)]Glc Xyl(b1-4)Xyl(b1-4)Xyl(b1-4)Xyl3Ac(b1-4)Xyl(b1-4)Xyl(b1-4)[GlcA(a1-2)]Xyl(b1-4)Xyl Xyl(b1-4)Xyl(b1-4)Xyl(b1-4)Xyl3Ac(b1-4)Xyl(b1-4)Xyl(b1-4)[GlcA(a1-2)]Xyl3Ac(b1-4)Xyl Xyl(b1-4)Xyl(b1-4)Xyl(b1-4)Xyl3Ac(b1-4)Xyl(b1-4)Xyl(b1-4)[GlcA4Me(a1-2)]Xyl(b1-4)Xyl Xyl(b1-4)Xyl(b1-4)Xyl(b1-4)Xyl3Ac(b1-4)Xyl(b1-4)Xyl(b1-4)[GlcA4Me(a1-2)]Xyl3Ac(b1-4)Xyl Xyl(b1-4)Xyl(b1-4)[GlcA(a1-2)]Xyl(b1-4)Xyl(b1-4)Xyl(b1-4)Xyl(b1-4)Xyl Xyl(b1-4)[GlcAOMe(a1-2)]Xyl(b1-4)Xyl(b1-4)Xyl(b1-4)Xyl Xyl2Ac3Ac4Ac(b1-3)Ara Xyl4Ac(b1-3)Ara XylOMe(b1-2)[RhaOMe(a1-6)]GlcOMe(b1-2)GlcOMe-ol XylOMe(b1-3)XylOMe(b1-2)[RhaOMe(a1-6)]GlcOMe(b1-2)GlcOMe-ol XylOMe(b1-4)RhaOMe(a1-2)D-FucOMe XylOMe(b1-4)RhaOMe(a1-2)[RhaOMe(a1-6)]GlcOMe XylOMe(b1-4)RhaOMe(a1-2)[RhaOMe(a1-6)]GlcOMe-ol Xylf(b1-2)Xyl(b1-3)[Rha(b1-2)Rha(b1-4)]Xyl [Araf(a1-3)Gal(b1-3)Gal(b1-6)]Gal(b1-3)Gal [Araf(a1-3)Gal(b1-6)]Gal(b1-3)Gal [Gal(a1-4)Gal(a1-6)]Man(b1-4)[Gal(a1-6)]Man(b1-4)[Man(b1-4)Man(b1-4)Man(b1-4)Gal(a1-6)]Man(b1-2)[Gal(a1-6)]Man(b1-2)[Gal(a1-4)Gal(a1-6)]Man(b1-4)Man [Gal(a1-6)]Man(b1-4)Man [Gal(a1-6)]Man(b1-4)Man(b1-4)Man [Gal(a1-6)]Man(b1-4)Man(b1-4)Man(b1-4)Man(b1-4)Man [Gal(a1-6)]Man(b1-4)[Gal(a1-6)]Man(b1-4)Man(b1-4)Man [Gal(a1-6)]Man(b1-4)[Gal(a1-6)]Man(b1-4)[Gal(a1-6)]Man(b1-4)Man [Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)Glc [Gal(b1-3)Gal(b1-6)[Araf(a1-3)]Gal(b1-6)]Gal(b1-3)Gal [Gal(b1-3)Gal(b1-6)]Gal(b1-3)Gal [Gal(b1-6)Gal(b1-6)Gal(b1-6)]Gal(b1-3)Gal [Gal(b1-6)Gal(b1-6)]Gal(b1-3)Gal [Gal(b1-6)]Gal(b1-3)Gal(b1-3)Gal(b1-3)Gal(b1-3)Gal(b1-3)Gal(b1-3)Gal [Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Araf(a1-2)Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)Glc [Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Araf(a1-2)Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)Glc [Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Araf(a1-2)Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)Glc [Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Araf(a1-2)Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)Glc [Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Gal(b1-5)Araf(a1-5)Araf(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)Glc [Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)Glc [Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Gal(b1-5)Araf(a1-5)Araf(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)Glc [Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Gal(b1-5)Araf(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)Glc [Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Fuc(a1-2)Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)Glc [Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Gal(b1-5)Araf(a1-5)Araf(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)Glc [Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Gal(b1-5)Araf(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)Glc [Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Gal(b1-5)Araf(a1-5)Araf(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)Glc [Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Gal(b1-5)Araf(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)Glc [Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Gal(b1-5)Araf(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Gal(b1-2)Xyl(a1-6)]Glc(b1-4)Glc(b1-4)Glc [Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)[Gal(b1-5)Araf(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)[Xyl(a1-6)]Glc(b1-4)Glc(b1-4)Glc
Family                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                        
Fabaceae 1 4 1 3 1 1 1 0 1 3 1 1 1 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 1 1 1 1 2 1 1 1 1 4 2 1 1 2 2 7 7 4 4 4 4 4 2 8 4 2 5 4 2 2 1 1 1 1 0 1 1 3 1 1 2 1 1 1 1 2 5 1 1 2 2 1 1 1 1 2 1 1 1 1 1 3 2 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 1 0 1 1 1 1 1 1 3 1 1 1 1 1 2 2 1 3 1 5 0 0 1 3 1 1 1 2 0 0 0 0 0 0 2 1 1 4 1 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 1 2 0 1 1 1 1 5 1 1 0 0 0 0 1 0 0 0 0 0 0 1 3 2 0 0 0 1 1 4 6 1 1 1 1 3 4 2 1 1 1 4 1 1 1 1 1 0 0 0 1 1 1 1 1 1 1 1 7 2 5 1 2 1 1 1 1 1 1 1 2 1 5 1 1 1 1 1 3 1 1 1 1 4 1 1 1 1 1 5 1 11 2 1 1 1 1 1 1 1 1 1 2 1 1 1 4 6 4 4 4 1 1 5 4 1 4 1 1 0 1 1 1 7 1 1 2 3 23 6 7 0 1 9 3 4 1 3 1 1 1 1 3 3 2 1 1 1 1 1 0 2 1 1 1 1 1 1 1 1 1 1 1 0 1 2 0 1 1 1 1 2 1 1 2 1 2 2 1 1 1 1 2 1 1 1 1 1 2 1 1 1 1 2 1 1 1 1 1 7 1 1 1 2 3 1 1 1 1 1 1 1 1 1 0 1 1 1 1 1 3 16 18 2 1 2 1 9 13 2 1 1 3 2 1 0 0 0 0 0 0 0 2 1 1 1 1 1 1 1 1 1 1 0 1 1 0 1 1 1 4 1 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
Fagaceae 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Polygalaceae 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 1 1 1 0 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 1 1 2 1 2 2 1 2 1 1 1 2 0 0 0 0 0 1 1 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 1 1 1 1 1 1 0 0 0 0 0 1 1 1 1 1 2 2 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 0 0 1 1 1 1 1 1 1 1 0 0 1 0 0 0 0 1 0 0 1 1 1 1 0 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
Quillajaceae 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

enforce_class

def enforce_class(
    glycan:str, # Glycan in IUPAC-condensed nomenclature
    glycan_class:str, # Glycan class (O, N, free, or lipid)
    conf:float | None=None, # Prediction confidence to override class
    extra_thresh:float=0.3, # Threshold to override class
)->bool: # True if glycan is in glycan class

Determines whether glycan belongs to a specified class

enforce_class("Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc", "O")
False

IUPAC_to_SMILES

def IUPAC_to_SMILES(
    glycan_list:str | list[str], # List of IUPAC-condensed glycans or single glycan
)->list[str]: # List of corresponding SMILES strings; empty for anything without a defined structure

Convert list of IUPAC-condensed glycans to isomeric SMILES

IUPAC_to_SMILES(['Neu5Ac(a2-3)Gal(b1-4)Glc'])
['OC1O[C@H](CO)[C@@H](O[C@@H]2O[C@H](CO)[C@H](O)[C@H](O[C@]3(C(=O)O)C[C@H](O)[C@@H](NC(C)=O)[C@H]([C@H](O)[C@H](O)CO)O3)[C@H]2O)[C@H](O)[C@H]1O']

canonicalize_composition

def canonicalize_composition(
    comp:str | dict[str, int], # Composition in Hex5HexNAc4Fuc1Neu5Ac2 or H5N4F1A2 format, or as a dict (e.g., {'Hex': 5})
    as_string:bool=False, # Whether to return canonical shorthand string (e.g., "H5N4F1A2") instead of dictionary
    strict:bool=False, # Whether to raise a ValueError for any residue that is no glycowork composition component with a mass (e.g., a peptide modification like 'Oxidation', or 'Galb' from a sequence); without it, an IUPAC-condensed sequence is read as its composition
    string_format:str='shorthand', # String written with as_string: 'shorthand' (H5N4F1A2), 'byonic' (HexNAc(4)Hex(5)Fuc(1)NeuAc(2), the glycan database format of Byonic and MSFragger-Glyco), or 'glycresoft' ({Fuc:1; Hex:5; HexNAc:4; Neu5Ac:2})
)->dict[str, int] | str: # Dictionary of monosaccharide:count, or canonical string if as_string

Converts composition from any common format to standardized dictionary or canonical shorthand string

print(canonicalize_composition("HexNAc2Hex1Fuc3Neu5Ac1"))
print(canonicalize_composition("N2H1F3A1"))
print(canonicalize_composition("HexNAc(4)Hex(5)Fuc(1)NeuAc(2) % 2350.8303", as_string = True))  # Byonic/MSFragger-Glyco
print(canonicalize_composition({'Hex': 5, 'HexNAc': 4}, as_string = True))
print(canonicalize_composition("H5N4F1A2", as_string = True, string_format = "byonic"))  # written back for a Byonic/MSFragger-Glyco glycan database
{'HexNAc': 2, 'Hex': 1, 'dHex': 3, 'Neu5Ac': 1}
{'HexNAc': 2, 'Hex': 1, 'dHex': 3, 'Neu5Ac': 1}
H5N4F1A2
H5N4
HexNAc(4)Hex(5)Fuc(1)NeuAc(2)

canonicalize_iupac

def canonicalize_iupac(
    glycan:str | int | list[str] | pandas.Series, # Glycan sequence in any supported format, or a list, tuple, array, Series, or Index of them
)->str | list[str] | pandas.Series: # Standardized IUPAC-condensed format; a list for a list/tuple/array, a Series/Index with the same index for a Series/Index

Convert glycan(s) from IUPAC-extended, LinearCode, GlycoCT, WURCS, Oxford, GLYCAM, GlycoWorkBench, pGlyco, StrucGP, CSDB-linear, KCF, SMILES, LINUCS, GlyConnect IDs, and GlyTouCanIDs to standardized IUPAC-condensed format

print(canonicalize_iupac("NeuAc?1-36SGalb1-4GlcNACb1-6(Fuc?1-2Galb1-4GlcNacb1-3Galb1-3)GalNAc-sp3"))
print(canonicalize_iupac("WURCS=2.0/5,11,10/[a2122h-1b_1-5_2*NCC/3=O][a1122h-1b_1-5][a1122h-1a_1-5][a2112h-1b_1-5][a1221m-1a_1-5]/1-1-2-3-1-4-3-1-4-5-5/a4-b1_a6-k1_b4-c1_c3-d1_c6-g1_d2-e1_e4-f1_g2-h1_h4-i1_i2-j1"))
print(canonicalize_iupac("Ma3(Ma6)Mb4GNb4GN;N"))
print(canonicalize_iupac("α-D-Manp-(1→3)[α-D-Manp-(1→6)]-β-D-Manp-(1→4)-β-D-GlcpNAc-(1→4)-β-D-GlcpNAc-(1→"))
print(canonicalize_iupac("""RES
1b:b-dgal-HEX-1:5
2s:n-acetyl
3b:b-dgal-HEX-1:5
4b:b-dglc-HEX-1:5
5b:b-dgal-HEX-1:5
6b:a-dglc-HEX-1:5
7b:b-dgal-HEX-1:5
8b:a-lgal-HEX-1:5|6:d
9b:a-dgal-HEX-1:5
10s:n-acetyl
11s:n-acetyl
12b:b-dglc-HEX-1:5
13b:b-dgal-HEX-1:5
14b:a-lgal-HEX-1:5|6:d
15b:a-lgal-HEX-1:5|6:d
16s:n-acetyl
17s:n-acetyl
18b:b-dgal-HEX-1:5
LIN
1:1d(2+1)2n
2:1o(3+1)3d
3:3o(3+1)4d
4:4o(-1+1)5d
5:5o(-1+1)6d
6:6o(-1+1)7d
7:7o(2+1)8d
8:7o(3+1)9d
9:9d(2+1)10n
10:6d(2+1)11n
11:5o(-1+1)12d
12:12o(-1+1)13d
13:13o(2+1)14d
14:12o(-1+1)15d
15:12d(2+1)16n
16:4d(2+1)17n
17:1o(6+1)18d
"""))
print(canonicalize_iupac("[][b-D-GlcpNAc]{[(4+1)][b-D-GlcpNAc]{[(4+1)][b-D-Manp]{[(3+1)][a-D-Manp]{}[(6+1)][a-D-Manp]{}}}}"))  # LINUCS
print(canonicalize_iupac("A2B2C1D1E2F1fedD1E2edcbB5ba"))  # StrucGP structure code
Fuc(a1-2)Gal(b1-4)GlcNAc(b1-3)Gal(b1-3)[Neu5Ac(a2-3)Gal6S(b1-4)GlcNAc(b1-6)]GalNAc
Fuc(a1-2)Gal(b1-4)GlcNAc(b1-2)Man(a1-6)[Gal(b1-4)GlcNAc(b1-2)Man(a1-3)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc
Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc
Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc
Fuc(a1-2)[GalNAc(a1-3)]Gal(b1-?)GlcNAc(a1-?)[Fuc(a1-2)Gal(b1-?)[Fuc(a1-?)]GlcNAc(b1-?)]Gal(b1-?)GlcNAc(b1-3)Gal(b1-3)[Gal(b1-6)]GalNAc
Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)GlcNAc
Gal(b1-3/4)GlcNAc(b1-2)Man(a1-3)[GlcNAc(b1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc

glycan_to_iupac_extended

def glycan_to_iupac_extended(
    glycan:str, # Glycan in IUPAC-condensed format
)->str: # Glycan in IUPAC-extended format as GlyTouCan writes it, e.g., β-D-Galp-(1→4)-?-D-GlcpNAc-(1→

Convert glycan from IUPAC-condensed to IUPAC-extended format, which states anomer, absolute configuration, and ring size of every monosaccharide

glycan_to_iupac_extended("Neu5Ac(a2-3)Gal(b1-4)[Fuc(a1-3)]GlcNAc")
'α-D-Neup5Ac-(2→3)-β-D-Galp-(1→4)-[α-L-Fucp-(1→3)]-?-D-GlcpNAc-(1→'

glycan_to_glycoct

def glycan_to_glycoct(
    glycan:str, # Glycan in IUPAC-condensed format
)->str: # Glycan in GlycoCT condensed format

Convert glycan from IUPAC-condensed to GlycoCT condensed format, e.g., for GlycoWorkbench, glypy, or GlyTouCan registration

print(glycan_to_glycoct("Neu5Ac(a2-3)Gal(b1-4)[Fuc(a1-3)]GlcNAc"))
RES
1b:x-dglc-HEX-1:5
2s:n-acetyl
3b:b-dgal-HEX-1:5
4b:a-dgro-dgal-NON-2:6|1:a|2:keto|3:d
5s:n-acetyl
6b:a-lgal-HEX-1:5|6:d
LIN
1:1d(2+1)2n
2:1o(4+1)3d
3:3o(3+2)4d
4:4d(5+1)5n
5:1o(3+1)6d

glycan_to_oxford

def glycan_to_oxford(
    glycan:str, # N-glycan in IUPAC-condensed format
)->str: # N-glycan in Oxford nomenclature, e.g., FA2G2S(6)2

Convert N-glycan from IUPAC-condensed to Oxford nomenclature (core fucose, antennae, bisecting GlcNAc, galactoses, sialic acids), which keeps the residue counts but not every linkage or arm

glycan_to_oxford("Neu5Ac(a2-6)Gal(b1-4)GlcNAc(b1-2)Man(a1-3)[Gal(b1-4)GlcNAc(b1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc")
'FA2G2S(6)1'

glycan_to_glycam

def glycan_to_glycam(
    glycan:str, # Glycan in IUPAC-condensed format, with every linkage defined
    reducing_anomer:str='b', # Anomer of the reducing-end residue ('a' or 'b'), which IUPAC-condensed leaves open but GLYCAM needs to build a structure; e.g., 'a' for mucin-type O-glycans
)->str: # Glycan in GLYCAM condensed nomenclature, e.g., DGalpb1-4DGlcpNAcb1-OH

Convert glycan from IUPAC-condensed to GLYCAM condensed nomenclature, e.g., as input for the GLYCAM-Web carbohydrate builder

glycan_to_glycam("Neu5Ac(a2-3)Gal(b1-3)GalNAc", reducing_anomer = "a")  # mucin-type O-glycans start with an alpha-linked GalNAc
'DNeup5Aca2-3DGalpb1-3DGalpNAca1-OH'

get_possible_linkages

def get_possible_linkages(
    wildcard:str, # Pattern to match, ? can be wildcard
    linkage_list:list[str]={'a2-11', 'b1-2', 'b2-7', '?1-6', '1-6', 'a1-11', 'b1-1', 'b1-4', '?1-?', '?2-3', 'a1-7', 'a1-8', 'a2-9', 'a1-3', 'b2-5', 'b1-6', 'a2-?', 'a2-2', '1-4', 'a1-9', 'a2-7', 'a1-4', 'a1-5', '?1-2', 'a2-1', 'a1-2', 'a1-?', 'a2-8', 'b1-3', 'a2-4', 'b1-?', 'b2-8', '?1-3', '?2-?', 'b1-9', 'a2-6', 'b2-6', 'b1-7', 'b2-1', 'a1-1', 'a1-6', 'b2-3', '?1-4', 'a2-5', 'b1-8', '?2-8', 'b2-2', 'a2-3', 'b2-4', 'b1-5', '?2-6'}, # List of linkages to search
)->frozenset[str]: # Matching linkages

Retrieves all linkages that match a given wildcard pattern

get_possible_linkages("a1-?")
frozenset({'a1-1',
           'a1-2',
           'a1-3',
           'a1-4',
           'a1-5',
           'a1-6',
           'a1-7',
           'a1-8',
           'a1-9',
           'a1-?'})

get_possible_monosaccharides

def get_possible_monosaccharides(
    wildcard:str, # Monosaccharide type; options: Hex, HexNAc, dHex, Sia, HexA, Pen, HexOS, HexNAcOS
)->frozenset[str]: # Matching monosaccharides

Retrieves all matching common monosaccharides of a type

get_possible_monosaccharides("HexNAc")
frozenset({'GalNAc', 'GlcNAc', 'HexNAc', 'ManNAc'})

equal_repeats

def equal_repeats(
    r1:str, # First glycan sequence
    r2:str, # Second glycan sequence
)->bool: # True if repeats are shifted versions

Check whether two repeat units could stem from the same repeating structure

equal_repeats("Fuc2S3S(a1-3)Fuc2S(a1-4)Fuc2S3S", "Fuc2S(a1-4)Fuc2S3S(a1-3)Fuc2S")
True

get_class

def get_class(
    glycan:str, # Glycan in IUPAC-condensed nomenclature
)->str: # Glycan class (repeat, O, N, free, lipid, lipid/free, or empty)

Determines glycan class

get_class("Gal(b1-4)GlcNAc(b1-2)Man(a1-3)[Gal(b1-4)GlcNAc(b1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc")
'N'

query

for interacting with the databases contained in glycowork, delivering insights for sequences of interest

get_insight

def get_insight(
    glycan:str, # Glycan in any format canonicalize_iupac reads (IUPAC-condensed/extended, WURCS, GlyTouCan ID, ...)
    motifs:pandas.DataFrame | None=None, # DataFrame of glycan motifs; default:motif_list
)->None: # Prints glycan meta-information

Print meta-information about a glycan

print("Test get_insight with 'Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc'")
get_insight('Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc')
Test get_insight with 'Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc'
Let's get rolling! Give us a few moments to crunch some numbers.

This glycan occurs in the following species: ['Acanthocheilonema_viteae', 'Adeno-associated_dependoparvovirusA', 'Aedes_aegypti', 'Angiostrongylus_cantonensis', 'Anopheles_gambiae', 'Antheraea_pernyi', 'Apis_mellifera', 'Ascaris_suum', 'Autographa_californica_nucleopolyhedrovirus', 'AvianInfluenzaA_Virus', 'Bombus_ignitus', 'Bombyx_mori', 'Bos_taurus', 'Brugia_malayi', 'Caenorhabditis_elegans', 'Cardicola_forsteri', 'Cooperia_onchophora', 'Cornu_aspersum', 'Crassostrea_gigas', 'Crassostrea_virginica', 'Cricetulus_griseus', 'Danio_rerio', 'Dictyocaulus_viviparus', 'Dirofilaria_immitis', 'Drosophila_melanogaster', 'Fasciola_hepatica', 'Gallus_gallus', 'Glossina_morsitans', 'Haemonchus_contortus', 'Haliotis_tuberculata', 'Heligmosomoides_polygyrus', 'Helix_lucorum', 'Homo_sapiens', 'HumanImmunoDeficiency_Virus', 'Hylesia_metabus', 'Hypsibius_exemplaris', 'Lutzomyia_longipalpis', 'Lymantria_dispar', 'Macaca_mulatta', 'Mamestra_brassicae', 'Megathura_crenulata', 'Mus_musculus', 'Nilaparvata_lugens', 'Oesophagostomum_dentatum', 'Onchocerca_volvulus', 'Onchocerca_volvulus', 'Ophiactis_savignyi', 'Opisthorchis_viverrini', 'Ostrea_edulis', 'Ovis_aries', 'Pan_troglodytes', 'Pristionchus_pacificus', 'Ramazzottius_varieornatus', 'Rattus_norvegicus', 'Schistosoma_mansoni', 'SemlikiForest_Virus', 'Spodoptera_frugiperda', 'Sus_scrofa', 'Tick_borne_encephalitis_virus', 'Tribolium_castaneum', 'Trichinella_spiralis', 'Trichoplusia_ni', 'Trichuris_suis', 'Tropidolaemus_subannulatus', 'Volvarina_rubella', 'undetermined', 'unidentified_influenza_virus']

Puh, that's quite a lot! Here are the phyla of those species: ['Arthropoda', 'Artverviricota', 'Chordata', 'Cossaviricota', 'Echinodermata', 'Kitrinoviricota', 'Mollusca', 'Negarnaviricota', 'Nematoda', 'Platyhelminthes', 'Tardigrada', 'Virus']

This glycan contains the following motifs: ['Chitobiose', 'Trimannosylcore', 'core_fucose']

This is the GlyTouCan ID for this glycan: G63041RA

This glycan has been reported to be expressed in: ['2A3_cell_line', 'A549_cell_line', 'AML_193_cell_line', 'C10_cell_line', 'CHOK1_cell_line', 'CHOS_cell_line', 'COLO_205_cell_line', 'COLO_320_cell_line', 'CRL_1620_cell_line', 'Caco_2_cell_line', 'Cal-27_cell_line', 'Cervicovaginal_Secretion', 'Co_115_cell_line', 'EOL_1_cell_line', 'FaDu_cell_line', 'HCT_15_cell_line', 'HCT_8_cell_line', 'HEK293_cell_line', 'HEL92_1_7_cell_line', 'HEL_cell_line', 'HL_60_cell_line', 'HT_29_cell_line', 'KG_1_cell_line', 'KG_1a_cell_line', 'KM12_cell_line', 'Kasumi_1_cell_line', 'LS174T_cell_line', 'LS180_cell_line', 'LS411N_cell_line', 'LoVo_cell_line', 'MDA_MB_231BR_cell_line', 'ME_1_cell_line', 'ML_1_cell_line', 'MOLM_13_cell_line', 'MOLM_14_cell_line', 'MV4_11_cell_line', 'M_07e_cell_line', 'NB_4_cell_line', 'NS0_cell_line', 'OCI_AML2_cell_line', 'OCI_AML3_cell_line', 'PLB_985_cell_line', 'RKO_cell_line', 'SCC-9_cell_line', 'SCC_25_cell_line', 'SW1116_cell_line', 'SW1398_cell_line', 'SW1463_cell_line', 'SW480_cell_line', 'SW48_cell_line', 'SW620_cell_line', 'SW948_cell_line', 'T84_cell_line', 'TF_1_cell_line', 'THP_1_cell_line', 'U_937_cell_line', 'VU-147T_cell_line', 'WiDr_cell_line', 'alveolus_of_lung', 'brain', 'brain', 'cerebellar_cortex', 'cerebellar_cortex', 'cerebellar_cortex', 'cerebellar_cortex', 'cerebellum', 'colon', 'cortex', 'digestive_tract', 'digestive_tract', 'forebrain', 'gills', 'gills', 'heart', 'heart', 'heart', 'hindbrain', 'hippocampal_formation', 'hippocampus', 'hippocampus', 'hippocampus', 'hippocampus', 'iPS1A_cell_line', 'iPS2A_cell_line', 'kidney', 'liver', 'liver', 'liver', 'lung', 'mantle', 'mantle', 'metastatic_pancreatic_ductal_adenocarcinoma', 'milk', 'mucus', 'muscle_of_leg', 'nerve_ending', 'ovary', 'pancreas', 'placenta', 'prefrontal_cortex', 'prefrontal_cortex', 'prefrontal_cortex', 'prefrontal_cortex', 'primary_pancreatic_ductal_adenocarcinoma', 'prostate_gland', 'seminal_fluid', 'striatum', 'striatum', 'striatum', 'striatum', 'testicle', 'testis', 'trachea', 'urine', 'urothelium']

This glycan has been reported to be dysregulated in (disease, direction, sample): [('REM_sleep_behavior_disorder', 'down', 'serum'), ('benign_breast_tumor_tissues_vs_para_carcinoma_tissues', 'up', 'breast'), ('cystic_fibrosis', 'up', 'sputum'), ('female_breast_cancer', 'up', 'breast'), ('female_breast_cancer', 'up', 'cell_line'), ('prostate_cancer', 'up', 'prostate_cancer_biopsy'), ('thyroid_gland_papillary_carcinoma', 'up', 'serum'), ('urinary_bladder_cancer', 'down', 'urine')]

That's all we can do for you at this point!

glytoucan_to_glycan

def glytoucan_to_glycan(
    ids:str | list[str], # GlyTouCan ID(s) or glycan(s)
    revert:bool=False, # Whether to map glycans to IDs; default:False
    verbose:bool=True, # Whether to print missing entries; default:True
)->str | list[str]: # Glycan(s) or ID(s), a single string for a single string input

Convert between GlyTouCan IDs and IUPAC-condensed glycans

glytoucan_to_glycan(['G63041RA'])
['Man(a1-3)[Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc']

smiles

convert glycan sequences into isomeric SMILES and back, without needing a cheminformatics toolkit

glycan_to_smiles

def glycan_to_smiles(
    glycan:str | networkx.classes.digraph.DiGraph, # Glycan in IUPAC-condensed format or as a networkx graph
    mapping:bool=False, # Also return, for every atom of the SMILES, the graph node it came from
    strict:bool=False, # Raise on an unknown linkage or modification position instead of taking the lowest free one
)->str | tuple: # Isomeric SMILES, or (SMILES, atom-to-node tuple) when mapping

Convert a glycan in IUPAC-condensed format into an isomeric SMILES string

glycan_to_smiles("Neu5Ac9Ac(a2-3)Gal6S(b1-3)[Neu5Gc(a2-6)]GalNAc")
'OC1O[C@H](CO[C@]2(C(=O)O)C[C@H](O)[C@@H](NC(=O)CO)[C@H]([C@H](O)[C@H](O)CO)O2)[C@H](O)[C@H](O[C@@H]2O[C@H](COS(=O)(=O)O)[C@H](O)[C@H](O[C@]3(C(=O)O)C[C@H](O)[C@@H](NC(C)=O)[C@H]([C@H](O)[C@H](O)COC(C)=O)O3)[C@H]2O)[C@H]1NC(C)=O'

smiles_to_iupac

def smiles_to_iupac(
    smiles:str, # SMILES string of a glycan
    strict:bool=False, # Raise on a substituent that cannot be named instead of dropping it
)->str: # Glycan in IUPAC-condensed format

Read a glycan’s SMILES back into IUPAC-condensed format, the inverse of glycan_to_smiles

smiles_to_iupac('OC1O[C@H](CO[C@]2(C(=O)O)C[C@H](O)[C@@H](NC(=O)CO)[C@H]([C@H](O)[C@H](O)CO)O2)[C@H](O)[C@H](O[C@@H]2O[C@H](COS(=O)(=O)O)[C@H](O)[C@H](O[C@]3(C(=O)O)C[C@H](O)[C@@H](NC(C)=O)[C@H]([C@H](O)[C@H](O)COC(C)=O)O3)[C@H]2O)[C@H]1NC(C)=O')
'Neu5Ac9Ac(a2-3)Gal6S(b1-3)[Neu5Gc(a2-6)]GalNAc'

parse_smiles

def parse_smiles(
    smiles:str, # SMILES string
)->tuple: # Atoms as (element, charge, chirality), bonds as (first, second, order), rings as tuples of bond indices, and each atom's neighbors in the order the string writes them

Read a SMILES into atoms, bonds, rings, and written neighbor order, which is what the chirality tags refer to

glycan_to_molecule

def glycan_to_molecule(
    glycan:str | networkx.classes.digraph.DiGraph, # Glycan in IUPAC-condensed format or as a networkx graph
    strict:bool=False, # Raise on an unknown linkage or modification position instead of taking the lowest free one
)->glycowork.motif.smiles.Molecule: # Atoms, bonds, rings, and the graph node every atom and bond came from

Build a glycan’s molecular graph without a cheminformatics toolkit, keeping every atom’s monosaccharide of origin

regex

for performing regular expression-like searches in glycans, very powerful to find complicated motifs

get_match

def get_match(
    pattern:str | list[str], # Expression or pre-compiled pattern; e.g., "Hex-HexNAc-([Hex|Fuc]){1,2}-HexNAc"
    glycan:str | networkx.classes.digraph.DiGraph, # Glycan string or graph
    return_matches:bool=True, # Whether to return matches vs boolean
)->bool | list[str]: # Match results

Find matches for glyco-regular expression in glycan

Glyco-regular expressions

A glyco-regular expression is read from the non-reducing end towards the reducing end, with - between residues meaning “any linkage”. get_match returns the matching substructures as IUPAC-condensed strings (return_matches = False gives a boolean), get_match_batch runs one pattern over many glycans, and motif_to_regex turns an existing motif into a pattern to start from.

Residues

syntax meaning
Gal, GlcNAc, … that exact monosaccharide
Hex, HexNAc, dHex, Sia, HexA, Pen, HexOS, HexNAcOS wildcard classes
. or Monosaccharide any monosaccharide
[Gal\|Man\|Fuc] one of these
[^Gal] or !Gal any monosaccharide that is not Gal

Linkages

syntax meaning
- any linkage
Mana6 Man(a1-6)
Galb3/4 Gal(b1-3/4)
.b3 any monosaccharide, b1-3-linked
Kdo(a2-4) write it out in full when the anomeric carbon is not 1

Quantifiers (append ? for the lazy form, which takes the fewest occurrences)

syntax meaning
{2} / {1,2} / {1,} / {,1} exactly / between / at least / at most
* / + / ? zero or more / one or more / zero or one

Branches, positions, context

syntax meaning
([Fuca3]){1} a branch hanging off the same residue as the preceding chunk; several can follow each other, and a branch may hold a chain such as ([Siaa3-Galb3]){1}
([!GlcNAcb6]){1} asserts that no such branch is present
^ / $ / % non-reducing terminus / reducing end / internal (has a child and is not the root)
(?<=Xyl-) / (?<!Xyl-) must / must not be preceded (leafward), not part of the match
(?=-HexNAc) / (?!-HexNAc) must / must not be followed (rootward), not part of the match

What comes back: only maximal matches, one per starting residue, ordered by length and then by position. A match contained in another is not reported separately.

N = "Neu5Ac(a2-3)Gal(b1-4)GlcNAc(b1-2)Man(a1-3)[Gal(b1-4)[Fuc(a1-3)]GlcNAc(b1-2)Man(a1-6)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc"
O = "Neu5Ac(a2-3)Gal(b1-3)[GlcNAc(b1-6)]GalNAc"
print(get_match("Hex-HexNAc", N))                    # ['Gal(b1-4)GlcNAc', 'Gal(b1-4)GlcNAc', 'Man(b1-4)GlcNAc']
print(get_match("Galb3/4-GlcNAc", N))                # linkage shorthand, 3 or 4
print(get_match("[Gal|Man]-HexNAc", N))              # alternatives
print(get_match("^Hex-HexNAc", N))                   # only the uncapped antenna
print(get_match(".-HexNAc$", N))                     # anchored at the reducing end
print(get_match("Hex%", O))                          # internal hexoses only
print(get_match("[Hex-HexNAc]{2,}", "Gal(b1-4)GlcNAc(b1-3)Gal(b1-4)GlcNAc(b1-3)Gal(b1-4)GlcNAc"))  # poly-LacNAc, whole run
print(get_match("Galb4-([Fuca3]){1}-GlcNAc", N))     # Lewis X via a branch group
print(get_match("Fuc-([^Gal])+-GlcNAc", "Fuc(a1-3)[GlcNAc(b1-4)]GlcNAc"))                          # negated branch
print(get_match("Galb3-([!GlcNAcb6]){1}-GalNAc", O), get_match("Galb3-([!GlcNAcb6]){1}-GalNAc", "Gal(b1-3)GalNAc"))  # core 1 but not core 2
print(get_match("(?<!Neu5Ac-)Galb4-GlcNAc", N))      # LacNAc that is not sialylated
print(motif_to_regex("Gal(b1-4)[Fuc(a1-3)]GlcNAc"))  # 'Fuca3-([Galb4]){1}-GlcNAc'
print(get_match_batch("Galb4-([Fuca3]){1}-GlcNAc", [N, O], return_matches = False))                # [True, False]
explain_match("Hex-HexNAc-([dHex]){1}-Hex", N)       # what each chunk compiled to, and where it occurs
['Gal(b1-4)GlcNAc', 'Gal(b1-4)GlcNAc', 'Man(b1-4)GlcNAc']
['Gal(b1-4)GlcNAc', 'Gal(b1-4)GlcNAc']
['Gal(b1-4)GlcNAc', 'Gal(b1-4)GlcNAc', 'Man(b1-4)GlcNAc']
['Gal(b1-4)GlcNAc']
['GlcNAc(b1-4)GlcNAc', 'Fuc(a1-6)GlcNAc']
['Gal']
['Gal(b1-4)GlcNAc(b1-3)Gal(b1-4)GlcNAc(b1-3)Gal(b1-4)GlcNAc']
['Fuc(a1-3)[Gal(b1-4)]GlcNAc']
['Fuc(a1-3)[GlcNAc(b1-4)]GlcNAc']
[] ['Gal(b1-3)GalNAc']
['Gal(b1-4)GlcNAc']
Fuca3-([Galb4]){1}-GlcNAc
[True, False]
chunk matches occurrences branch position lookaround hits_in_glycan
0 Hex-HexNAc Hex(?1-?)HexNAc 1-1 False 3
1 ([dHex]){1} dHex 1-1 True 2
2 Hex Hex 1-1 False 5

For interested users, we here compile a selection of regular expression patterns that we find useful in our own work:

  • Lewis or sialyl-Lewis structures:
    pattern = r”[Sia]{,1}-[Gal|GalOS]{1}-([Fuc]){1}-[GlcNAc|GlcNAc6S]{1}”
  • Blood groups:
    pattern = r”Fuc-([Gal|GalNAc])?-Gal-GlcNAc”
  • a1-6 branch in N-glycans:
    pattern = r”[Sia]{,1}-[Hex|HexNAc]{,1}-([dHex]){,1}-[Man|GlcNAc]{1}-([.-.|.]){,1}-Mana6(?=-Manb4-GlcNAc)”
  • b1-6 branch in O-glycans (from core 2/4/6):
    pattern = r”[Sia|dHex]{,1}-[Hex|HexNAc]{,1}-([dHex]){,1}-.b6(?=-GalNAc)”
  • b1-3 branch in O-glycans (from core 1/2):
    pattern = r”[Sia]{,1}-[.]{,1}-([dHex]){,1}-.b3(?=-GalNAc)”

get_match_batch

def get_match_batch(
    pattern:str | list[str], # Expression or pre-compiled pattern; e.g., "Hex-HexNAc-([Hex|Fuc]){1,2}-HexNAc"
    glycan_list:str | networkx.classes.digraph.DiGraph | list[str | networkx.classes.digraph.DiGraph], # Glycan(s) as strings or graphs
    return_matches:bool=True, # Whether to return matches vs boolean
)->list[bool] | list[list[str]]: # Match results for each glycan

Find glyco-regular expression matches in list of glycans

explain_match

def explain_match(
    pattern:str, # Glyco-regular expression, e.g., "Hex-HexNAc-([Hex|Fuc]){1,2}-HexNAc"
    glycan:str | networkx.classes.digraph.DiGraph, # Glycan string or graph
)->pandas.DataFrame: # One row per pattern chunk, with what it compiled to and how often it occurs

Show what each chunk of a glyco-regular expression means and where it does or does not occur in a glycan

motif_to_regex

def motif_to_regex(
    motif:str, # Glycan in IUPAC-condensed
)->str: # Regular expression

Convert glycan motif to regular expression pattern

motif_to_regex("Fuc(a1-3)[Gal(b1-4)]GlcNAc(b1-?)")
'Fuca3-([Galb4]){1}-GlcNAcb?'

tokenization

helper functions to map m/z–>composition, composition–>structure, structure–>motif, and more

string_to_labels

def string_to_labels(
    character_string:str, # String to tokenize
    libr:dict[str, int] | None=None, # Dictionary mapping characters to indices
)->list[int]: # List of character indices

Tokenize word by indexing characters in library

string_to_labels(['Man','a1-3','Man','a1-6','Man'])
[None, None, None, None, None]

pad_sequence

def pad_sequence(
    seq:list[int], # Sequence to pad
    max_length:int, # Target length
    pad_label:int | None=None, # Padding token value
    libr:dict[str, int] | None=None, # Character library
)->list[int]: # Padded sequence

Pad sequences to same length using padding token

pad_sequence(string_to_labels(['Man','a1-3','Man','a1-6','Man']), 7)
[None, None, None, None, None, 25, 25]

stemify_glycan

def stemify_glycan(
    glycan:str, # Glycan in IUPAC-condensed format
    stem_lib:dict[str, str] | None=None, # Modified to core monosaccharide mapping; default:created from lib
    libr:dict[str, int] | None=None, # Glycoletter to index mapping
)->str: # Stemmed glycan string

Remove modifications from all monosaccharides in glycan

stemify_glycan("Neu5Ac9Ac(a2-3)Gal6S(b1-3)[Neu5Ac(a2-6)]GalNAc")
'Neu5Ac(a2-3)Gal(b1-3)[Neu5Ac(a2-6)]GalNAc'

stemify_dataset

def stemify_dataset(
    df:pandas.DataFrame, # DataFrame with glycan column
    stem_lib:dict[str, str] | None=None, # Modified to core monosaccharide mapping; default:created from lib
    libr:dict[str, int] | None=None, # Glycoletter to index mapping
    glycan_col_name:str='glycan', # Column name for glycans
    rarity_filter:int=1, # Modified monosaccharides occurring more often than this keep their modification
)->pandas.DataFrame: # DataFrame with stemified glycans

Remove monosaccharide modifications from all glycans in dataset

mask_rare_glycoletters

def mask_rare_glycoletters(
    glycans:list[str], # List of IUPAC-condensed glycans
    thresh_monosaccharides:int | None=None, # Threshold for rare monosaccharides (default: 0.001*len(glycans))
    thresh_linkages:int | None=None, # Threshold for rare linkages (default: 0.03*len(glycans))
)->list[str]: # List of glycans with masked rare elements

Mask rare monosaccharides and linkages in glycans

mz_to_composition

def mz_to_composition(
    mz_value:float, # m/z value from mass spec
    max_charge:int=-2, # Signed charge ceiling: sign sets ion mode (negative/positive), magnitude the highest charge state z considered
    mass_value:str='monoisotopic', # Mass type: monoisotopic/average
    modification:str | None=None, # Reducing end modification: reduced/2AA/2AB/procainamide
    sample_prep:str='underivatized', # Sample preparation method: underivatized/permethylated/peracetylated
    mass_tolerance:float=0.5, # Mass tolerance for matching
    tolerance_unit:str='Da', # Unit of mass_tolerance: "Da" (absolute) or "ppm"
    kingdom:str='Animalia', # Taxonomic kingdom filter for choosing a subset of glycans to consider
    glycan_class:str='all', # Glycan class: N/O/lipid/free/all
    df_use:pandas.DataFrame | None=None, # Custom glycan database
    filter_out:str | set[str] | None=None, # Monosaccharide(s) to ignore during composition finding
    deprioritized:str | set[str] | None={'PCho', 'HexA', 'Me'}, # Monosaccharide(s) to use only as fallback if no other composition matches
    extras:str | list[str]=[], # Additional operation(s): 'adduct' also considers the adduct_ions
    adduct:str | None=None, # Chemical formula of a neutral adduct added to every candidate's mass, e.g., "C2H4O2"
    adduct_ions:str | list[str] | None=None, # Adduct ion(s) considered with extras = ['adduct'], as named in mz_to_composition.csv (e.g., 'Na+', 'K+', 'NH4+', 'Acetate', 'Formate'); default: Acetate (negative) or Na+ (positive)
    mass_tag:float | None=None, # Mass in Da a reducing-end label adds to the free glycan, i.e., the label minus O (e.g., 121.0528 for 2AA, 120.0687 for 2AB, 219.1735 for procainamide), added to every candidate's neutral mass
    sialic_derivatization:str | tuple[str, str] | None=None, # Linkage-specific sialic acid derivatization ('EE', 'EEA', 'ME', 'DMA', 'SALSA', or a pair of SIALIC_PRODUCTS for a2-6 and a2-3/8/9 sialic acids); compositions then come with linkage-resolved sialic acids (Neu5Ac_a2-6, Neu5Ac_a2-3)
    return_all:bool=False, # Whether to return every composition and ion within mass_tolerance, as GlycoMod lists them, instead of the first hit
)->list[dict[str, int]] | pandas.DataFrame: # List of matching compositions; with return_all a DataFrame of every match (composition, ion, theoretical mz, error_ppm) in Occam order, deprioritized ones last

Map m/z value to matching monosaccharide composition

mz_to_composition(665.4, glycan_class='O', filter_out={'Kdn', 'P', 'HexA', 'Pen', 'HexN', 'Me', 'PCho', 'PEtN'},
                    modification = "reduced")
[{'Hex': 2, 'HexNAc': 2, 'Neu5Ac': 2}]

match_composition_relaxed

def match_composition_relaxed(
    composition:dict[str, int] | str, # Composition as a dict (e.g., {"dHex": 1, "Hex": 1, "HexNAc": 1}) or in any format canonicalize_composition reads (e.g., "H1N1F1")
    glycan_class:str='N', # Glycan class: N/O/lipid/free/all
    kingdom:str='Animalia', # Taxonomic kingdom filter for choosing a subset of glycans to consider
    df_use:pandas.DataFrame | None=None, # Custom glycan database
)->list[str]: # List of matching glycans

Map coarse-grained composition to matching glycans

match_composition_relaxed({"Hex":3, "HexNAc":2, "dHex":1}, glycan_class = 'O')
['Fuc(a1-2)[Gal(a1-3)]Gal(b1-4)GlcNAc(b1-6)[Gal(b1-3)]GalNAc',
 'Fuc(a1-2)[Gal(a1-3)]Gal(b1-3)[Gal(b1-4)GlcNAc(b1-6)]GalNAc',
 'Gal(b1-4)Gal(b1-4)[Fuc(a1-3)]GlcNAc(b1-6)[Gal(b1-3)]GalNAc',
 'Gal(?1-3/4)Gal(b1-3/4)[Fuc(a1-3/4)]GlcNAc(b1-6)[Gal(b1-3)]GalNAc',
 'Gal(b1-4)Gal(b1-4)[Fuc(a1-3)]GlcNAc(b1-3)Gal(b1-3)GalNAc',
 'Fuc(a1-2)Gal(b1-3/4)GlcNAc(b1-3)Gal(b1-3)[Gal(b1-6)]GalNAc',
 'Fuc(a1-2)Gal(b1-4)GlcNAc(b1-6)[Gal(?1-?)Gal(b1-3)]GalNAc',
 'Fuc(a1-2)[Gal(a1-3)]Gal(b1-3)GlcNAc(b1-3)Gal(b1-3)GalNAc',
 'Fuc(a1-2)[Gal(a1-3)]Gal(b1-4)GlcNAc(?1-3/4)Gal(b1-3)GalNAc',
 'Fuc(a1-3)[Gal(b1-4)]GlcNAc(b1-3)Gal(b1-4)GlcNAc(b1-3)Gal',
 'Gal(?1-?)Gal(b1-4)GlcNAc(b1-6)[Fuc(a1-2)Gal(b1-3)]GalNAc',
 'Fuc(a1-2)Gal(b1-4)GlcNAc(b1-6)[Gal(a1-3)Gal(b1-3)]GalNAc',
 'Gal(a1-3)Gal(b1-4)GlcNAc(b1-6)[Fuc(a1-2)Gal(b1-3)]GalNAc',
 'Fuc(a1-2)Gal(b1-3)Gal(b1-3)GlcNAc(b1-6)[Gal(b1-3)]GalNAc',
 'Fuc(a1-2)Gal(b1-3)Gal(b1-3)[Gal(b1-4)GlcNAc(b1-6)]GalNAc',
 'Gal(b1-4)Gal(b1-3)[Fuc(a1-3)[Gal(b1-4)]GlcNAc(b1-6)]GalNAc',
 'Fuc(a1-2)Gal(?1-?)Gal(b1-3/4)GlcNAc(b1-6)[Gal(b1-3)]GalNAc',
 'Gal(a1-3)GalNAc(a1-3)[Fuc(a1-2)]Gal(b1-3)Gal(b1-3)GalNAc',
 'Man(a1-6)Glc(a1-4)GlcNAc(b1-4)[Fuc(a1-2)]Gal(b1-3)GalNAc',
 'Man(a1-6)Glc(b1-4)GlcNAc(b1-4)[Fuc(a1-2)]Gal(b1-3)GalNAc',
 'Fuc(a1-2)Gal(b1-3)GlcNAc(b1-3)Gal(b1-4)GlcNAc(b1-?)Man',
 'Gal(b1-2)Gal(a1-3)[Fuc(a1-2)]Gal(b1-3)[GlcNAc(b1-6)]GalNAc',
 'Fuc(a1-2)Gal(a1-3)Gal(a1-4)Gal(b1-3)[GlcNAc(b1-6)]GalNAc',
 'Gal(b1-4)[Fuc(a1-3)]GlcNAc(b1-6)[Gal(b1-3)]Gal(b1-3)GalNAc',
 'Fuc(a1-3)[Gal(b1-4)]GlcNAc(b1-?)Gal(b1-6)[Gal(b1-3)]GalNAc']

condense_composition_matching

def condense_composition_matching(
    matched_composition:list[str], # List of matching glycans
)->list[str]: # Minimal list of representative glycans

Find minimum set of glycans characterizing matched composition

match_comp = match_composition_relaxed({'Hex':1, 'HexNAc':1, 'Neu5Ac':1}, glycan_class = 'O')
print(match_comp)
condense_composition_matching(match_comp)
['Neu5Ac(a2-3)Gal(b1-3)GalNAc', 'Gal(b1-3)[Neu5Ac(a2-6)]GalNAc', '{Neu5Ac(a2-3/6)}Gal(b1-3)GalNAc', 'Neu5Ac(a2-3)[GalNAc(b1-4)]Gal', 'Gal(a1-3)[Neu5Ac(a2-6)]GalNAc', 'Neu5Ac(a2-3/6)Gal(b1-3)GalNAc', 'Neu5Ac(a2-6)Gal(b1-3)GalNAc', 'Gal(?1-3)[Neu5Ac(a2-6)]GalNAc', 'Neu5Ac(a2-3/6)Gal(?1-3)GalNAc', 'Neu5Ac(a2-?)Hex(?1-?)GalNAc', 'Neu5Ac(a2-3)Gal(?1-?)GalNAc', 'Neu5Ac(a2-3/6)GalNAc(a1-6)Gal', 'Neu5Ac(a2-6)Gal(a1-3)GalNAc', 'Gal(b1-4)[Neu5Ac(a2-6)]GalNAc', 'Neu5Ac(a2-3)GalNAc(b1-3)Gal']
['Neu5Ac(a2-3)Gal(b1-3)GalNAc',
 'Gal(b1-3)[Neu5Ac(a2-6)]GalNAc',
 '{Neu5Ac(a2-3/6)}Gal(b1-3)GalNAc',
 'Neu5Ac(a2-3)[GalNAc(b1-4)]Gal',
 'Gal(a1-3)[Neu5Ac(a2-6)]GalNAc',
 'Neu5Ac(a2-6)Gal(b1-3)GalNAc',
 'Neu5Ac(a2-3/6)GalNAc(a1-6)Gal',
 'Neu5Ac(a2-6)Gal(a1-3)GalNAc',
 'Gal(b1-4)[Neu5Ac(a2-6)]GalNAc',
 'Neu5Ac(a2-3)GalNAc(b1-3)Gal']

mz_to_structures

def mz_to_structures(
    mz_list:float | list[float], # Precursor mass(es)
    glycan_class:str, # Glycan class: N/O/lipid/free/all
    kingdom:str='Animalia', # Taxonomic kingdom filter for choosing a subset of glycans to consider
    abundances:pandas.DataFrame | None=None, # Sample abundances matrix
    max_charge:int=-2, # Signed charge ceiling: sign sets ion mode (negative/positive), magnitude the highest charge state z considered
    mass_value:str='monoisotopic', # Mass type: monoisotopic/average
    sample_prep:str='underivatized', # Sample prep: underivatized/permethylated/peracetylated
    mass_tolerance:float=0.5, # Mass tolerance for matching
    tolerance_unit:str='Da', # Unit of mass_tolerance: "Da" (absolute) or "ppm"
    modification:str | None=None, # Reducing end modification: reduced/2AA/2AB/procainamide
    df_use:pandas.DataFrame | None=None, # Custom glycan database
    filter_out:str | set[str] | None=None, # Monosaccharide(s) to ignore
    deprioritized:str | set[str] | None={'PCho', 'HexA', 'Me'}, # Monosaccharide(s) to use only as fallback if no other composition matches
    verbose:bool=False, # Whether to print non-matching compositions
    mass_tag:float | None=None, # Mass in Da a reducing-end label adds to the free glycan, i.e., the label minus O (e.g., 121.0528 for 2AA), added to every candidate's neutral mass
    sialic_derivatization:str | tuple[str, str] | None=None, # Linkage-specific sialic acid derivatization ('EE', 'EEA', 'ME', 'DMA', 'SALSA', or a pair of SIALIC_PRODUCTS); the mass then gives each sialic acid's linkage, and only structures with these linkages are returned, unknown ones filled in where the mass decides them
)->pandas.DataFrame | list: # DataFrame of structures x intensities or empty list

Map precursor masses to structures, supporting accompanying relative intensities

mz_to_structures([674.29], glycan_class = 'O')
glycan abundance
0 Fuc(a1-2)Gal(b1-4)[Fuc(a1-3)]GlcNAc 0

compositions_to_structures

def compositions_to_structures(
    composition_list:str | dict[str, int] | list[str | dict[str, int]], # Composition(s) like {'Hex': 1, 'HexNAc': 1} or 'H1N1'
    glycan_class:str='N', # Glycan class: N/O/lipid/free/all
    kingdom:str='Animalia', # Taxonomic kingdom filter for choosing a subset of glycans to consider
    abundances:pandas.DataFrame | None=None, # Sample abundances matrix
    df_use:pandas.DataFrame | None=None, # Custom glycan database
    verbose:bool=False, # Whether to print non-matching compositions
)->pandas.DataFrame: # DataFrame of structures x intensities

Map compositions to structures, supporting accompanying relative intensities

compositions_to_structures([{'Neu5Ac': 2, 'Hex': 1, 'HexNAc': 1}], glycan_class = 'O')
glycan abundance
0 Neu5Ac(a2-3)Gal(b1-3)[Neu5Ac(a2-6)]GalNAc 0
1 Neu5Ac(a2-8)Neu5Ac(a2-6)[Gal(b1-3)]GalNAc 0
2 Neu5Ac(a2-3)[Neu5Ac(a2-6)]Gal(b1-3)GalNAc 0
3 Neu5Ac(a2-3)Gal(b1-4)[Neu5Ac(a2-6)]GalNAc 0 
compositions_to_structures(["H1N1A2"], glycan_class = 'O')
glycan abundance
0 Neu5Ac(a2-3)Gal(b1-3)[Neu5Ac(a2-6)]GalNAc 0
1 Neu5Ac(a2-8)Neu5Ac(a2-6)[Gal(b1-3)]GalNAc 0
2 Neu5Ac(a2-3)[Neu5Ac(a2-6)]Gal(b1-3)GalNAc 0
3 Neu5Ac(a2-3)Gal(b1-4)[Neu5Ac(a2-6)]GalNAc 0

structure_to_basic

def structure_to_basic(
    glycan:str, # Glycan in IUPAC-condensed format
)->str: # Base topology string

Convert glycan structure to base topology

structure_to_basic("Neu5Ac(a2-3)Gal6S(b1-3)[Neu5Ac(a2-6)]GalNAc")
'Neu5Ac(?1-?)HexOS(?1-?)[Neu5Ac(?1-?)]HexNAc'

glycan_to_composition

def glycan_to_composition(
    glycan:str, # Glycan in IUPAC-condensed format
    stem_libr:dict[str, str] | None=None, # Modified to core monosaccharide mapping; default: created from lib
    sialic_linkage:bool=False, # Whether to count Neu5Ac, Neu5Gc, and Kdn by linkage, as a linkage-specific derivatization tells them apart: a2-6 (Neu5Ac_a2-6) or lactone-forming a2-3/8/9 (Neu5Ac_a2-3); one of unknown linkage (a2-3/6) stays Neu5Ac
)->dict[str, int]: # Dictionary of monosaccharide counts; empty if a residue or substituent has no composition key

Map glycan to its composition

glycan_to_composition("Neu5Ac(a2-3)Gal6S(b1-3)[Neu5Ac(a2-6)]GalNAc")
{'Hex': 1, 'HexNAc': 1, 'Neu5Ac': 2, 'S': 1}

glycan_to_mass

def glycan_to_mass(
    glycan:str | dict[str, int], # Glycan in IUPAC-condensed format, or a composition (dict, or a string canonicalize_composition reads, such as "H5N4F1")
    mass_value:str='monoisotopic', # Mass type: monoisotopic/average
    sample_prep:str='underivatized', # Sample prep: underivatized/permethylated/peracetylated
    stem_libr:dict[str, str] | None=None, # Modified to core monosaccharide mapping
    adduct:str | float | None=None, # Chemical formula of adduct (e.g., "C2H4O2") OR its exact mass in Da
    modification:str | None=None, # Reducing end modification: reduced/2AA/2AB/procainamide; a glycan ending in an alditol (-ol) is reduced already
    peptide:str | None=None, # Peptide carrying the glycan, in one-letter code with optional signed mass or formula deltas after a residue (e.g., 'EEQYNSTYR', 'NC[+C2H3NO]SK'), to get the glycopeptide mass
    sialic_derivatization:str | tuple[str, str] | None=None, # Linkage-specific sialic acid derivatization of an underivatized sample ('EE', 'EEA', 'ME', 'DMA', 'SALSA', or a pair of SIALIC_PRODUCTS for a2-6 and a2-3/8/9 sialic acids); each sialic acid then needs a known linkage class, a2-3/6 or a2-? raises
)->float: # Theoretical mass

Calculate theoretical mass from glycan or composition, optionally as a glycopeptide

glycan_to_mass("Neu5Ac(a2-3)Gal6S(b1-3)[Neu5Ac(a2-6)]GalNAc")
1045.290418

composition_to_mass

def composition_to_mass(
    dict_comp_in:dict[str, int] | str, # Composition dictionary of monosaccharide:count, or a string in any format canonicalize_composition reads (e.g., "H5N4F1")
    mass_value:str='monoisotopic', # Mass type: monoisotopic/average
    sample_prep:str='underivatized', # Sample prep: underivatized/permethylated/peracetylated
    adduct:str | float | None=None, # Chemical formula of adduct (e.g., "C2H4O2") OR its exact mass in Da
    modification:str | None=None, # Reducing end modification: reduced/2AA/2AB/procainamide
    peptide:str | None=None, # Peptide carrying the glycan(s), in one-letter code with optional signed mass or formula deltas after a residue (e.g., 'EEQYNSTYR', 'NC[+C2H3NO]SK' for carbamidomethylation, 'M[+15.9949]NGTK'), to get the glycopeptide mass; several glycans on one peptide are passed as their total composition
    sialic_derivatization:str | tuple[str, str] | None=None, # Linkage-specific sialic acid derivatization of an underivatized sample, a key of SIALIC_DERIVATIZATIONS ('EE', 'EEA', 'ME', 'DMA', 'SALSA') or a pair of SIALIC_PRODUCTS (a2-6, a2-3/8/9); every sialic acid then needs its linkage ('H5N4E1L1', {'Neu5Ac_a2-6': 1, 'Neu5Ac_a2-3': 1}), and uronic acids take the a2-6 product; peptide carboxyls are left to peptide's deltas
)->float: # Theoretical mass

Calculate theoretical mass from composition, optionally as a glycopeptide

print(composition_to_mass({'Neu5Ac': 2, 'Hex': 1, 'HexNAc': 1, 'S': 1}))
print(composition_to_mass('H5N4F1', peptide = 'EEQYNSTYR'))  # glycopeptide mass (IgG1 Fc glycopeptide with G2F)
1045.290418
2957.144267

calculate_adduct_mass

def calculate_adduct_mass(
    formula:str, # Chemical formula of adduct (e.g., "C2H4O2", "-H2O", "+Na")
    mass_value:str='monoisotopic', # Mass type: monoisotopic/average
    enforce_sign:bool=False, # If True, returns 0 for unsigned formulas
)->float: # Formula mass

Calculate mass of adduct from chemical formula, including signed formulas

calculate_adduct_mass("C2H4O2")
60.02113

get_formula

def get_formula(
    glycan:str | dict[str, int], # Glycan in any nomenclature canonicalize_iupac reads, composition (dict, or a string canonicalize_composition reads, e.g., "H5N4F1"), or elemental formula (e.g., 'C62H104N4O46')
    sample_prep:str='underivatized', # Sample prep: underivatized/permethylated/peracetylated
    stem_libr:dict[str, str] | None=None, # Modified to core monosaccharide mapping
    adduct:str | None=None, # Chemical formula of a neutral adduct (e.g., "C2H4O2"); a signed formula ("-H2O") removes atoms
    modification:str | None=None, # Reducing end modification: reduced/2AA/2AB/procainamide; a glycan ending in an alditol (-ol) is reduced already
    peptide:str | None=None, # Peptide carrying the glycan(s), in one-letter code with optional signed formula deltas after a residue (e.g., 'EEQYNSTYR', 'NC[+C2H3NO]SK'); several glycans on one peptide are passed as their total composition
    sialic_derivatization:str | tuple[str, str] | None=None, # Linkage-specific sialic acid derivatization of an underivatized sample ('EE', 'EEA', 'ME', 'DMA', 'SALSA', or a pair of SIALIC_PRODUCTS for a2-6 and a2-3/8/9 sialic acids); every sialic acid then needs its linkage ('H5N4E1L1', 'Neu5Ac(a2-6)...')
)->str: # Elemental formula in Hill order (C, H, then alphabetical), e.g., 'C62H104N4O46' for H5N4

Elemental formula of a glycan, composition, or formula: the molecule glycan_to_mass weighs

print(get_formula("Neu5Ac(a2-3)Gal(b1-3)GalNAc"))
print(get_formula("H1N1A1"))
C25H42N2O19
C25H42N2O19

get_unique_topologies

def get_unique_topologies(
    composition:dict[str, int] | str, # Composition dictionary of monosaccharide:count, or a string in any format canonicalize_composition reads (e.g., "H5N4F1")
    glycan_type:str, # Glycan class: N/O/lipid/free/repeat
    df_use:pandas.DataFrame | None=None, # Custom glycan database to use for mapping
    universal_replacers:dict[str, str] | None=None, # Base-to-specific monosaccharide mapping
    taxonomy_rank:str='Kingdom', # Taxonomic rank for filtering
    taxonomy_value:str='Animalia', # Value at taxonomy rank
)->list[str]: # List of unique base topologies

Get all observed unique base topologies for composition

get_unique_topologies({'HexNAc':2, 'Hex':1}, 'O', universal_replacers = {'dHex':'Fuc'})
['Hex(?1-?)[HexNAc(?1-?)]HexNAc',
 'Hex(?1-?)HexNAc(?1-?)HexNAc',
 'HexNAc(?1-?)Hex(?1-?)HexNAc',
 'HexNAc(?1-?)[HexNAc(?1-?)]Hex',
 'HexNAc(?1-?)HexNAc(?1-?)Hex']

get_random_glycan

def get_random_glycan(
    n:int=1, # How many random glycans to sample
    glycan_class:str='all', # Glycan class: N/O/lipid/free/repeat/all
    kingdom:str='Animalia', # Taxonomic kingdom filter for choosing a subset of glycans to consider
)->str | list[str]: # Returns a random glycan or list of glycans if n > 1

Sample random glycans from the SugarBase database

get_random_glycan()
'GlcA(b1-4)GalNAc(b1-4)GlcNAc(b1-4)GlcNAc(b1-2)Man(a1-6)[Fuc(a1-3)[GalNAc(b1-4)]GlcNAcOPCho(b1-2)Man(a1-3)]Man(b1-4)GlcNAc(b1-4)[Fuc(a1-6)]GlcNAc'