def pdb_mapgaps(mapgaps_profile, pdb_fname):
"""Align a PDB structure to a MAPGAPS profile.
Returns a tuple: (SeqRecord, list of aligned residue numbers)
"""
from biocma import cma
with read_pdb_seq(pdb_fname) as (seqfname, seqs):
call_quiet('run_gaps', mapgaps_profile, seqfname)
pdb_cma = cma.read(seqfname + '_aln.cma')
hits = {}
head_lengths = {}
for seq in pdb_cma['sequences']:
hits[seq['id']] = seq['seq']
head_lengths[seq['id']] = seq['head_len']
ref_id, ref_aln = choose_best_aligned(hits)
ref_record = SeqIO.to_dict(seqs)[ref_id]
offset = (ref_record.annotations['start'] + head_lengths[ref_id] - 1)
resnums, inserts = aln_resnums_inserts(ref_record, ref_aln, offset)
return ref_record, resnums, inserts
def test_get_inserts(self):
block = cma.read(EX_CMA)
inserts = utils.get_inserts(block)
self.assertEqual(len(inserts), len(block['sequences']))
fullseqs = SeqIO.to_dict(SeqIO.parse(EX_FASTA, 'fasta'))
for sequence, targets in (
(block['sequences'][1], ['n', 't', 'fyklyllkkydsntlfnv']),
(block['sequences'][-1], ['altkl', 'nkl',
'siptvgfskdgdrlqemykasvcsyteecqg',
'ndndgeylldge', 'eh', 'p',
'epecancneedknmsennhkkdskhkgdsnhksdsnhksdsnhksdsnhksgsnhksdcnhksgsnhksdsnhqsdcnhmsdhnhksdnnhksdsshksdsshksdsshksgsnhksdnnhksdsshksgsnhksdhnhksdsnhksdsnhknesnhknesnhknesnhknesnhknesnhkndsnhksdsnhmsdhnhksdnnhksdhnhmsdhnhksdnnhksdnnhmsdhnhksdnnhksdnnhksdnnhksdhnhmsdhnhksdnnhksdhnhksdsnhmsdhnhmsdhnhksdhnhksdhnhksdnnhksdsnhksdsnhksdhnhksdsnhmsdhnhmsdhnhksdhnhksdnnhksdsnhksdsnhksdhnhksdsnhmsdhnhmsdhnhmsdhnhksdhnhksdnnhksdsnhksdsnhksdsnhksdhnhksdhkhmsdnnhksdnnhksdhnhksdnnhksdhnhksdsnhksdsnhksdsnhksdsnhksdnnhksdhnhnsdsnhmsdhnhksdhnhksdhnhksdnnhksdnnhksdhnhksdhkknnnnnkdnknddnddsdasdavhediellesysdlnkfnemlteqln',
'vt', 'edtrv', 'pmythnl', 'g',
'sfqscqpcv', 'iirehiklkidnpfehlstitdqee',
'yfd', 'ra', 'fqlak'])):
full = fullseqs[sequence['id']]
ins_ranges = [str(full.seq)[start-1:end]
for start, end in inserts[sequence['id']]]
print sequence['id'], ins_ranges
self.assertEqual(len(ins_ranges), len(targets))
for ins, tgt in zip(ins_ranges, targets):
self.assertEqual(ins.lower(), tgt)
def pdb_mapgaps(mapgaps_profile, pdb_fname):
"""Align a PDB structure to a MAPGAPS profile.
Returns a tuple: (SeqRecord, list of aligned residue numbers)
"""
from biocma import cma
with read_pdb_seq(pdb_fname) as (seqfname, seqs):
subprocess.check_call(['run_gaps', mapgaps_profile, seqfname])
pdb_cma = cma.read(seqfname + '_aln.cma')
hits = {}
head_lengths = {}
for seq in pdb_cma['sequences']:
hits[seq['id']] = seq['seq']
head_lengths[seq['id']] = seq['head_len']
ref_id, ref_aln = choose_best_aligned(hits)
ref_record = SeqIO.to_dict(seqs)[ref_id]
offset = (ref_record.annotations['start']
+ head_lengths[ref_id]
- 1)
resnums, inserts = aln_resnums_inserts(ref_record, ref_aln, offset)
return ref_record, resnums, inserts
def test_read(self):
block = cma.read(EX_CMA)
self.assertEqual(len(block['sequences']), 24)
self.assertEqual(block['query_length'], block['sequences'][0]['length'])
graph.add((ruri, MSA.deleted_aln_pos, Literal(i)))
else:
graph.add((ruri, RDF.type, MSA.aligned_residue))
graph.add((ruri, MSA.aln_pos, Literal(i)))
if unquote(acc) in dedup_eqv[i]:
graph.add((ruri, MSA.native_pos, Literal(dedup_eqv[i][unquote(acc)])))
else:
print "shouldn't happen" #deletions taken care of above
graph.add((ruri, MSA.native_residue, Literal(r)))
bar.next()
bar.finish()
graph.serialize(destination=outfile, format='pretty-xml')
return
if __name__ == "__main__":
import argparse
parser = argparse.ArgumentParser(description = 'Populate the Multiple Sequence Alignment Ontology')
parser.add_argument('infile', metavar='infile', type=str, help='input aligned sequences')
parser.add_argument('name', metavar='name', type=str, help='Graph name')
parser.add_argument('--namespace', metavar='namespace', type=str, default='http://localhost/msaont#', help='Graph namespace')
parser.add_argument('-p', dest='prop', action='store', default=0.25, help='proportion of inserts allowed in an aligned residue')
parser.add_argument('-o', dest='outfile', action='store', default='out.rdf', help='outfile')
args = parser.parse_args()
ext = args.infile.split('.')[-1]
if ext != 'cma':
seqs = manipulate_fasta(args.infile, args.prop)
else:
seqs = cma.read(args.infile)
populate(seqs, args.name, args.outfile, args.namespace)