def _test_hypergeom(m, n, n1, n2):
R_pval = r.phyper(min(n1, n2), n1, n - n1, n2)[0] \
- r.phyper(m - 1, n1, n - n1, n2)[0]
f_pval = fisher.pvalue(*_fury_table(m, n, n1, n2)).right_tail
# at least to 10 sig figs
R_str = ('%.10f' % R_pval)
f_str = ('%.10f' % f_pval)
print 'R:', R_str, 'Fisher:', f_str
assert R_str == f_str
def _test_hypergeom(m, n, n1, n2):
R_pval = r.phyper(min(n1, n2), n1, n - n1, n2)[0] \
- r.phyper(m - 1, n1, n - n1, n2)[0]
f_pval = fisher.pvalue(*_fury_table(m, n, n1, n2)).right_tail
# at least to 10 sig figs
R_str = ('%.10f' % R_pval)
f_str = ('%.10f' % f_pval)
print 'R:', R_str, 'Fisher:', f_str
assert R_str == f_str
def test_enrichment(test_set, gene_set, universe):
"""
Tests the test_set for enrichment relative to the gene_set using the hypergeometric test.
Returns the probability of seeing at least this many draws in the test set that come from the gene set.
@arg test_set: The gene set we wish to test for enrichment.
@arg gene_set: The background set of interest, e.g. genes in a GO category.
@arg universe: The genes we have selected our gene set from.
"""
from rpy2.robjects import r
assert test_set <= universe
assert gene_set <= universe
white_drawn = len(test_set.intersection(gene_set))
white = len(gene_set)
black = len(universe) - len(gene_set)
draws = len(test_set)
#import IPython; IPython.Debugger.Pdb().set_trace()
kwargs = {'lower.tail': False}
logging.debug('%4d black; %4d white; %4d draws; %4d white draws', black,
white, draws, white_drawn)
return white_drawn, white, black, draws, r.phyper(white_drawn - 1, white,
black, draws,
**kwargs)[0]