def all_identities(source_ps, odict_fname, source_fasta, target_fasta,
target_id_dict_fname=None):
odict = load_odict(odict_fname)
if target_id_dict_fname is not None:
tid_dict = dict([(x[0],x[2]) for x in
ut.load_lot(target_id_dict_fname)])
odict = dict([(k,[tid_dict[v] for v in vs]) for k,vs in odict.items()])
dsource = load_seq_dict(source_fasta)
dtarget = load_seq_dict(target_fasta)
pairs = [(s, odict[s][0]) for s in source_ps if s in odict]
print "%s of %s with orthologs--getting identities" % (len(pairs),
len(source_ps))
idents = []
for s,t in pairs:
try:
ident = seqs.percent_identity(dsource[s], dtarget[t])
idents.append(ident)
except decorators.TimeoutError, ex:
print "timeout for %s %s" %(s,t), ex.args
except Exception, ex:
print "unknown error for %s %s" % (s,t), ex.args
def exported_diff(cy_basefile, cy_difffile, col_header, diff_ppis=None,
justids=False):
"""
Makes a new cy_ file labeling whether that interaction is also found in the
cy_difffile (or the diff_ppis--pass None for cy_difffile in that case).
"""
def cy_ppi_to_pair(p):
return (p[0].split('_')[1], p[1].split('_')[1])
if cy_difffile is not None:
pd_diff = pd.PairDict([cy_ppi_to_pair(p)
for p in ut.load_lot(cy_difffile)[1:]])
else:
pd_diff = pd.PairDict(diff_ppis)
header = ut.load_lol(cy_basefile)[0]
lines = ut.load_lol(cy_basefile)[1:]
if justids:
lines = [l[:2] for l in lines]
header = header[:2]
header += [col_header]
ut.write_tab_file([r + [pd_diff.contains(cy_ppi_to_pair(r))] for r in
lines], ut.pre_ext(cy_basefile, col_header), header=header)
def load_hpa_localization(fname='./enrichment_datasets/subcellular_location.csv'):
locs = ut.load_lot(fname, sep=",")[1:]
locs_clean = [[x.strip("\"") for x in l] for l in locs]
locs_filt = [l for l in locs_clean if l[4]=="Supportive"]
locs = [(l[0],l[1]) for l in locs_filt]
return locs