API Reference¶

`DiscoverMatrix`(events[, bg, strata])	Class to store a binary alteration matrix and the corresponding alteration probability estimates.
`pairwise_discover_test`(x[, g, alternative, …])	Perform many pairwise mutual exclusivity or co-occurrence tests.
`PairwiseDiscoverResult`(pvalues, qvalues, …)	Class to store the results of the pairwise_discover_test function.
`row_stack`(matrices)	Stack DiscoverMatrix objects row-wise.

Alteration matrices¶

class discover.DiscoverMatrix(events, bg=None, strata=None)¶

Class to store a binary alteration matrix and the corresponding alteration probability estimates.

Parameters

eventspandas.DataFrame: Binary alteration matrix.
bgpandas.DataFrame, optional: Matrix of alteration probabilities for events. Mainly for internal use only. Use at your own risk.
strataarray_like, optional: To perform a stratified DISCOVER test, this array should contain the strata. The length of this array must match the number of columns of events.

Attributes

discover.row_stack(matrices)¶

Stack DiscoverMatrix objects row-wise.

Parameters

matricessequence of DiscoverMatrix: Sequence containing the DiscoverMatrix objects to be stacked. The matrices must have the same number of columns, and the column names must match.

Returns

discover.pairwise_discover_test(x, g=None, alternative='less', fdr_method='DBH')¶

Perform many pairwise mutual exclusivity or co-occurrence tests.

Parameters

xDiscoverMatrix
garray_like, optional: An optional grouping vector for the rows of x. Pairs of rows within the same group are not tested.
alternative{‘less’, ‘greater’}, optional: If ‘less’, a mutual-exclusivity analysis is performed, if ‘greater’ a co-occurrence analysis.
fdr_method{‘DBH’, ‘BH’}, optional: The false discovery rate procedure used for multiple testing correction. If ‘DBH’, a Benjamini-Hochberg procedure adapted for discrete test statistics is used. If ‘BH’, the standard Benjamini-Hochberg procedure is used. The latter is much faster, but also more conservative than the discrete version.

Returns

resultPairwiseDiscoverResult: An object containing the test results for all pairwise combinations.

class discover.pairwise.PairwiseDiscoverResult(pvalues, qvalues, pi0, alternative, fdr_method)¶

Class to store the results of the pairwise_discover_test function.

Attributes

pvalues, qvaluespandas.DataFrame: Matrices containing the pairwise DISCOVER test P values and FDR-corrected Q values. The P value corresponding to a gene pair (gene1, gene2) is stored in either pvalues.ix[gene1, gene2] or pvalues.ix[gene2, gene1]. The other entry will contain nan. Q values are stored in the same way.
pi0float: Estimate of the proportion of true null hypotheses.
alternative{‘less’, ‘greater’}: If ‘less’, these results relate to mutual exclusivity, if ‘greater’ to co-occurrence.
fdr_method{‘DBH’, ‘BH’}: The method used to estimate these results’ q-values.

significant_pairs(q_threshold=0.01)¶

Return the gene pairs significant at a specified maximum false discovery rate.

Parameters

q_thresholdfloat, optional: The maximum false discovery rate (default 0.01) for test results included in the result.

Returns

resultpandas.DataFrame: DataFrame with significant gene pairs. The gene names are stored in the columns gene1 and gene2. The columns pvalue and qvalue contain the DISCOVER test P value and the FDR-corrected Q value.

discover.groupwise_discover_test(events, method='impurity')¶

Perform a groupwise mutual exclusivity test.

Parameters

eventsDiscoverMatrix: Matrix with rows corresponding to the genes in the gene set to be tested.
method{‘impurity’, ‘coverage’, ‘exclusivity’}: The mutual exclusivity statistic to estimate significance for.

Returns