def msgf2seq_file(filepath, fasta_file, msb_psms):
"""
msb_psms: set of spectid_peptidesequence
"""
def parse_spec_pep_row(r):
# get spec_pep from _best file format
parsed = '_'.join(r[0].split('.')[:2] + [r[4]])
#print parsed
return parsed
usedir,fin = os.path.split(filepath)
# Get the sample filename from the first item of the third line
fout = next(it.islice(ut.load_tab_file(filepath),2,3))[0].split('.')[0]
in_gen = ut.load_tab_file(filepath)
in_gen.next(); in_gen.next() # skip 2 lines
p2g = seqs.prots2genes(fasta_file)
g2p = ut.dict_inverse(p2g)
fout = os.path.join(usedir, '.'.join([fout, fin.split('.')[-1] ,
'sequestformat']))
search = searches[filepath.split('.')[-1]]
print "Converting/filtering; Search:", search
output = (msgfbest2sequest_line(r,p2g, g2p, search) for r in in_gen
if parse_spec_pep_row(r) in msb_psms)
print "Writing", fout
ut.write_tab_file(output, fout)
return fout
def ensg_to_ensp_and_park(ppips):
dhpg = seqs.prots2genes('/Users/blakeweb/Dropbox/complex/data/sequences/canon/Hs.fasta')
dhgp = ut.dict_inverse(dhpg)
parkids = ut.load_lol('./orth_similarities/table.Hsapiens/Hsapiens_id.txt')
ppips_ensp = [dhgp[g] for g in ppips]
dg2park = dict([(x[2],x[0]) for x in parkids])
dp2park = dict([(x[1],x[0]) for x in parkids])
park_ppips_most = [dp2park[p] for p in ppips_ensp if p in dp2park]
ppips_ensp_rest = [p for p in ppips_ensp if p not in dp2park]
ppips_ensg_rest = [dhpg[p] for p in ppips_ensp_rest]
park_ppips_rest = [dg2park[p] for p in ppips_ensg_rest if p in dg2park]
park_ppips = park_ppips_most + park_ppips_rest
return park_ppips
def check(fasta, protq, do_convert):
p2g = seqs.prots2genes(fasta)
g2p = ut.dict_inverse(p2g)
fprots = el.load_elution(protq).prots
print "checking", ut.shortname(protq)
print "proteins: %s of %s" % (len([p for p in fprots if p in p2g]),
len(fprots))
ngenesfound = len([p for p in fprots if p in g2p])
print "genes: %s of %s" % (ngenesfound,
len(fprots))
if do_convert and ngenesfound < len(fprots):
print "converting prots to genes:", protq
seqs.elut_p2g(protq, p2g)
