def test_look_for_factor_pair():
print 'test_look_for_factor_pair'
all_pairs = biopsy.load(
'c:/data/remos/0/0.top_pairs',
display = 'pairs log odds'
)
sascha_factors = [
"T00140", # c-Myc human, H**o sapiens
"T00163", # CREB human, H**o sapiens
"T00368", # HNF-1alpha-A human, H**o sapiens
"T00594", # RelA-p65 human, H**o sapiens
"T00671", # p53 human, H**o sapiens
"T00759", # Sp1 human, H**o sapiens
"T00781", # TAF(II)250 human, H**o sapiens
"T03286", # HNF-6alpha human, H**o sapiens
"T03828", # HNF-4alpha human, H**o sapiens
]
sascha_pssms = set()
for factor in sascha_factors:
pssms_for_factor = biopsy.get_pssms_for_factor( factor )
print 'Pssms for %s: %s' % ( factor, " ".join( pssms_for_factor ) )
for pssm in pssms_for_factor:
sascha_pssms.add( pssm )
look_for_factor_pair(
'sascha',
all_pairs,
sascha_pssms,
sascha_pssms
)
def test_pair_go_analysis():
import rpy
rpy.r.library("Category")
rpy.r.library("biomaRt")
rpy.r.source('../../R/go_categorise.R')
remome_threshold = 100
analysis_file = 'c:/data/remos/%d/%d.analysis' \
% ( remome_threshold, remome_threshold )
pairs_file = 'C:/Data/ReMos/%d/%d.top_pairs' \
% ( remome_threshold, remome_threshold )
gene_universe_file = 'c:/Data/remos/%d/%d.uniq_genes' \
% ( remome_threshold, remome_threshold )
gene_universe = rpy.r.read_table( gene_universe_file )["V1"]
analysis = biopsy.Analysis.deserialise( analysis_file )
pairs = biopsy.load( pairs_file, display = 'pairs' )
biopsy.print_pairs_go_analysis(
pairs[:150],
gene_universe,
analysis
)
print 'done'
def test_saschas_pairs():
pair_dists = biopsy.load( 'saschas.pairs' )
for p, d in pair_dists.iteritems():
if ( p[0] == 'M01019' and p[1] == 'M01043' ) \
or ( p[1] == 'M01019' and p[0] == 'M01043' ):
print p
print d
def pair_log_odds_histogram():
pairs = biopsy.load(
'c:/data/remos/100/100.pairs_lor',
display = 'pairs log odds'
)
import pylab, math
pylab.hist(
[ p.log_odds_ratio for p in pairs ],
100
)
pylab.show()
def test_get_pairs_for_pssm():
pssm = 'M00736'
pairs = biopsy.load(
'c:/data/remos/100/100.top_pairs',
display = 'pairs log odds'
)
pssm_pairs = biopsy.get_pairs_for_pssm(
pairs,
pssm
)
print 'Got %d pairs for %s from total of %d' % \
( len(pssm_pairs), pssm, len(pairs) )
for p in pssm_pairs[:20]:
print p.binder_pair, p.log_odds_ratio
biopsy.write_pair_separation_histograms( pssm_pairs[:20], pssm )
def test_research_pair():
import biopsy.r_go, rpy
pair = ('M00293', 'R04602', False, True)
remome_threshold = 100
analysis_file = 'c:/data/remos/%d/%d.analysis' % ( remome_threshold, remome_threshold )
pairs_file = 'C:/Data/ReMos/%d/%d.top_pairs' % ( remome_threshold, remome_threshold )
gene_universe_file = 'c:/Data/remos/%d/%d.uniq_genes' % ( remome_threshold, remome_threshold )
gene_universe = rpy.r.read_table( gene_universe_file )["V1"]
analysis = biopsy.Analysis.deserialise( analysis_file )
pairs = biopsy.load( pairs_file, display = 'pairs' )
mart = biopsy.r_go.get_mart( "mmusculus_gene_ensembl" )
categs = biopsy.r_go.categorise_genes( mart, gene_universe )
biopsy.research_pair( analysis, mart, categs, pair )
