def run_msa(gene_dict, rs_data, problems):
# Next, get sequences and run alignments
counter = 0
matches = set()
aln_data = {}
for gene_sym, rs_ids in gene_dict.items():
counter += 1
#if counter >= 20:
# break
print("%s: %d of %d genes" % (gene_sym, counter, len(gene_dict)))
fasta_lines = []
# Get the main Uniprot sequence from the gene symbol
hgnc_id = hgnc_client.get_hgnc_id(gene_sym)
up_id_main = hgnc_client.get_uniprot_id(hgnc_id)
up_sequence = uniprot_client.get_sequence(up_id_main)
fasta_lines.append('>%s\n' % gene_sym)
fasta_lines.append('%s\n' % up_sequence)
# Now, iterate over the refseq ids and get the sequences
seq_ids = []
# The filenames to use if we do an alignment
in_file = 'aln/in/%s.fasta' % gene_sym
out_file = 'aln/out/%s.fasta' % gene_sym
# Iterate over the Refseq IDs
for rs_id in rs_ids:
seq_info = rs_data.get(rs_id)
if not seq_info:
problems.add((rs_id, 'no sequence in Refseq'))
continue
seq_ids.append(rs_id)
fasta_header, sequence = seq_info
fasta_lines.append('>%s\n%s\n' % (rs_id, sequence))
if sequence == up_sequence:
aln_data[rs_id] = (gene_sym, True, None)
else:
aln_data[rs_id] = (gene_sym, False, out_file)
if len(seq_ids) == 0:
continue
if len(seq_ids) == 1 and sequence == up_sequence:
print("\tAll sequences match, no alignment needed.")
continue
else:
# Write the fasta file
with open(in_file, 'wt') as f:
for line in fasta_lines:
f.write(line)
# Run the sequence alignment
print("\tRunning sequence alignment.")
subprocess.call(['./clustal-omega-1.2.3-macosx', '-i', in_file,
'-o', out_file, '--force'])
return aln_data
def test_get_sequence():
seq = uniprot_client.get_sequence('P00533')
assert len(seq) > 1000
assert unicode_strs(seq)
for site_tuple in sites:
s = site_tuple[1]
split_id = s.split('-')
if len(split_id) == 1 or split_id[1] == '1':
continue
else:
iso_sites.append(site_tuple)
iso_only = []
pos_valid_in_ref = []
mappable_in_ref = []
for site in iso_sites:
site_id, up_id, res, pos, motif = site
ref_iso = up_id.split('-')[0]
ref_seq = uniprot_client.get_sequence(ref_iso)
assert len(motif) == 13
ref_motif_start = ref_seq.find(motif)
if ref_motif_start == -1:
iso_only.append(site)
else:
ref_pos = str(ref_motif_start + 6)
if ref_pos == pos:
pos_valid_in_ref.append(site)
else:
mappable_in_ref.append(site)
"""
with open('ms_data/tubulin_sorted_site_list.txt', 'rt') as f:
for line in f.readlines():
gene_name, respos = line.strip().split('_')
res = respos[0]
def test_get_sequence():
seq = uniprot_client.get_sequence('P00533')
assert len(seq) > 1000
assert unicode_strs(seq)
def test_get_sequence():
seq = uniprot_client.get_sequence('P00533')
assert(len(seq) > 1000)