def findEnrichedAnnotationFreqs_motifLs(fg_ls, bg_ls, annotations, motifs, fasta_file, permutations):
""" Check to see if the given motifs are enriched in the fg,
as compared to the bg. Return motif {}, each entry
is [#of proteins with motif, |fg|, # of times bg
permutation is as good as fg]. Do X background
permutations.
@param fg_ls: {} of foreground genes
@param bg_ls: {} of backgound genes
@param annotations: protein 2 annotation;
@param motifs: {} of motifs you are checking for enrichment
@return: [#of proteins with motif, |fg|, # of times bg is as good as fg] for each motif
"""
fasta = utils_fasta.loadFASTA(fasta_file)
scores = scoreProteinLsFreqs_motifLs(fg_ls, annotations, motifs, fasta)
better_counts = {}
for motif in motifs.keys():
better_counts[motif] = 0
for i in xrange(permutations):
bck_scores = scoreBckgndFreqs_motifLs(len(fg_ls), bg_ls,
annotations, motifs, fasta)
for motif in motifs.keys():
if bck_scores[motif] >= scores[motif]:
better_counts[motif] += 1
for motif in motifs.keys():
scores[motif] = [ scores[motif], len(fg_ls), better_counts[motif] ]
return scores
else:
tempSeq = ''
for s in seq:
tempSeq = tempSeq + s
offset = 0
while match:
printResult(protein, elm, match, tempSeq, offset)
tempSeq = tempSeq[int(match.start())+1:]
offset += int( match.start() ) + 1
match = p.search(tempSeq)
req_args = ['pattern file',
'fasta file']
examples = ['../../Data/ELM/elm2pattern',
'../../Data/FASTA/Human/hprd.intr.fasta']
utils_scripting.checkStart(sys.argv, req_args, examples, len(req_args), True)
input_pattern_file = sys.argv[1]
elm2pattern = {}
f = open(input_pattern_file)
for line in f.xreadlines():
[elm, pattern] = map(string.strip, line.split())
elm2pattern[elm] = [re.compile(pattern), pattern]
f.close()
fasta = utils_fasta.loadFASTA(sys.argv[2])
for protein in fasta.keys():
matchSeq(protein, fasta[protein], elm2pattern)
import markov_chain, utils_fasta
chain = markov_chain.MarkovChain()
f = utils_fasta.loadFASTA('/home/perry/bioperry/Projects/Thesis/Data/FASTA/Human/hprd_new.intr.fasta')
for k in f:
chain.add(f[k])
for i in range(10):
print "".join(chain.random_output())
vp_suffix = sys.argv[2]
protein_ls_output_file = sys.argv[3]
new2old_file = sys.argv[4]
old2new_file = sys.argv[5]
def clean(seq):
""" Remove dashes and stars. """
return seq.replace('-', '').replace('*', '').lstrip().strip()
protein_ls = {}
new2old = {}
old2new = {}
old_fasta = utils_fasta.loadFASTA(sys.argv[1])
for protein in old_fasta.keys():
new_seq = clean(old_fasta[protein])
name = protein
if name.split('.')[-1] != vp_suffix:
name = name + '.' + vp_suffix
utils_fasta.prettyPrint(name, new_seq)
protein_ls[name] = True
new2old[name] = {}
old2new[name] = {}
new_index = 0
for old_index in xrange(len(old_fasta[protein])):
old_residue = old_fasta[protein][old_index]
if old_residue == '-' or old_residue == '*':
pass
else:
import utils_fasta, sys, utils, global_settings
from collections import defaultdict
aa = 0
for g in global_settings.GENOMES:
fasta = utils_fasta.loadFASTA('data/' + g + '.fa')
for p in fasta:
aa += len(fasta[p])
elm2count = defaultdict(utils.init_zero)
for g in global_settings.GENOMES:
with open('results/elmdict_' + g + '.txt') as f:
for line in f:
[elm, seq, count_st, freq] = line.strip().split('\t')
elm2count[elm] += int(count_st)
with open('results/all_elm_aa_freq', 'w') as f:
for elm in elm2count:
v = float(elm2count[elm])/float(aa)
f.write(elm + '\t' + str(v) + '\n')
