flotilla.visualize.network module¶

Visualize results from :py:mod:flotilla.compute.network

class flotilla.visualize.network.NetworkerViz(DataModel)[source]¶

Bases: flotilla.compute.network.Networker

draw_graph(n_pcs=5, use_pc_1=True, use_pc_2=True, use_pc_3=True, use_pc_4=True, degree_cut=2, cov_std_cut=1.8, weight_function='no_weight', featurewise=False, rpkms_not_events=False, feature_of_interest='RBFOX2', draw_labels=True, reduction_name=None, feature_ids=None, sample_ids=None, graph_file='', compare='', sample_id_to_color=None, label_to_color=None, label_to_marker=None, groupby=None, data_type=None)[source]¶

Draw the graph of similarities between samples or features

Parameters:

feature_ids : list of str, or None Feature ids to subset the data. If None, all features will be used. sample_ids : list of str, or None Sample ids to subset the data. If None, all features will be used. x_pc : str, optional Which component to use for the x-axis, default “pc_1” y_pc : y component for PCA, default “pc_2” n_pcs : int Number of components to use for cells’ covariance calculation cov_std_cut : float Covariance cutoff for edges use_pc{1-4} : bool Use these pcs in cov calculation (default True) degree_cut : int miniumum degree for a node to be included in graph display weight_function : [‘arctan’ | ‘sq’ | ‘abs’ | ‘arctan_sq’] weight function (arctan (arctan cov), sq (sq cov), abs (abs cov), arctan_sq (sqared arctan of cov)) gene_of_interest : str map a gradient representing this gene’s data onto nodes (ENSEMBL id or gene symbol)

Returns:

graph : networkx.Graph positions : (x,y) positions of nodes

draw_nonreduced_graph(degree_cut=2, cov_std_cut=1.8, wt_fun='abs', featurewise=False, rpkms_not_events=False, feature_of_interest='RBFOX2', draw_labels=True, feature_ids=None, group_id=None, graph_file='', compare='')[source]¶

Parameters:

feature_ids : list of str, or None Feature ids to subset the data. If None, all features will be used. sample_ids : list of str, or None Sample ids to subset the data. If None, all features will be used. x_pc : str x component for DataFramePCA, default “pc_1” y_pc : y component for DataFramePCA, default “pc_2” n_pcs : int??? n components to use for cells’ covariance calculation cov_std_cut : float?? covariance cutoff for edges use_pc{1-4} use these pcs in cov calculation (default True) degree_cut : int?? miniumum degree for a node to be included in graph display weight_function : [‘arctan’ | ‘sq’ | ‘abs’ | ‘arctan_sq’] weight function (arctan (arctan cov), sq (sq cov), abs (abs cov), arctan_sq (sqared arctan of cov)) gene_of_interest : str map a gradient representing this gene’s data onto nodes (ENSEMBL id or gene name???)

Returns:

#TODO: Mike please fill these in graph : networkx.Graph ??? positions : ??? ???