def run(
motif_csv, test_fasta, out_html, aa_probs,
variant='fixed',
alpha=1.0, beta=0.01):
log_odds_pssm = get_motif_log_odds(
motif_csv, aa_probs,
alpha=alpha, beta=beta,
variant=variant)
write_log_odds(log_odds_pssm, motif_csv.replace('.pssm', '').replace('.csv', '.pssm'))
n_position = len(log_odds_pssm.keys())
seqids, fastas = uniprot.read_fasta(test_fasta)
saved_scores = []
html = "<body>"
for seqid, entry in fastas.items():
# print "Sequence", seqid
test_seq = entry['sequence']
scores = score_full_sequence(log_odds_pssm, test_seq)
html += "<h1>%s</h1>" % entry['description']
html += make_html_for_seq_scores(test_seq, n_position, scores)
saved_scores.append({
'seqid': seqid,
'description': entry['description'],
'scores': scores,
'seq': test_seq
})
html += "</body>"
open(out_html, 'w').write(html)
titles = ['seqid', 'description', 'position', 'seq', 'score']
rows = [titles]
for entry in saved_scores:
n_seq = len(entry['seq'])
for i in range(0, n_seq - n_position):
score = entry['scores'][i]
pos = i + 1
if score > 0:
row = [
entry['seqid'],
entry['description'],
pos,
score,
entry['seq'][i:i+n_position]
]
rows.append(row)
datafile.write_csv(rows, out_html + '.csv')
def count_aa_in_fasta(fasta, counts_txt):
if os.path.isfile(counts_txt):
with open(counts_txt) as f:
return eval(f.read())
seqids, fastas = uniprot.read_fasta(fasta)
counts = {}
n = 0
for entry in fastas.values():
seq = entry['sequence']
for c in seq:
if c not in counts:
counts[c] = 0
counts[c] += 1
n += len(seq)
with open(counts_txt, 'w') as f:
f.write(repr(counts))
return counts
import os
import uniprot
import pprint
import sys
# Clean up caches
os.system('rm cache*')
# Example 1 - reading a fasta file
seqids, fastas = uniprot.read_fasta('example.fasta')
pprint.pprint(seqids, indent=2)
# Example 2 - map identifiers for RefSeq to Uniprot
seqids = "NP_000508.1 NP_001018081.3".split()
pairs = uniprot.batch_uniprot_id_mapping_pairs('P_REFSEQ_AC', 'ACC', seqids)
pprint.pprint(pairs, indent=2)
# Example 2 - get UniProt metadata
uniprot_seqids = [j for i, j in pairs]
uniprot_data = uniprot.batch_uniprot_metadata(uniprot_seqids, 'cache')
pprint.pprint(uniprot_data, indent=2)
# Example 3 - parse for isoforms in metadata
text = open('cache/metadata.0.txt').read()
uniprot_data = uniprot.parse_isoforms(text)
pprint.pprint(uniprot_data)
# Example 4 - chaining commands to map seqids
seqids = "EFG_MYCA1 YP_885981.1 ENSG00000196176 Q91ZU6-8".split()
uniprot_data = uniprot.get_metadata_with_some_seqid_conversions(
seqids, 'cache2')
fname = 'KL-A1_pep15.csv'
peptides = []
proteins = {}
for row in csv.DictReader(open(fname)):
seqid = row['Protein']
if seqid not in proteins:
proteins[seqid] = []
proteins[seqid].append(row)
row['matches'] = []
peptides.append(row)
seqids, fasta = uniprot.read_fasta('use.fasta')
for seqid, protein in proteins.items():
bare_seqid = clean_seqid(seqid)
full_sequence = fasta[bare_seqid]['sequence']
for i_peptide1, peptide1 in enumerate(protein):
seq1 = peptide1['Sequence']
i1 = full_sequence.find(seq1)
peptide1['Start'] = i1 + 1
peptide1['End'] = i1 + len(seq1)
for peptide2 in protein:
if peptide1 == peptide2:
continue
seq2 = peptide2['Sequence']
i2 = full_sequence.find(seq2)
match = {
import os
import uniprot
import pprint
# Clean up caches
os.system("rm *output* *cache*")
# Example 1 - reading a fasta file
seqids, fastas = uniprot.read_fasta("example.fasta")
pprint.pprint(seqids, indent=2)
# Example 2 - batch read identifier mappings with
# prespecified identifier types
seqids = """
NP_000508.1 NP_001018081.3
""".split()
pairs = uniprot.batch_uniprot_id_mapping_pairs("P_REFSEQ_AC", "ACC", seqids)
pprint.pprint(pairs, indent=2)
# Example 3 - sequential identifier mapping to UniProt
# identifiers using robust but slow method
('%s%d' % (cell_type, i_repeat), color, i_source))
print source_sets
protein = {}
for exp, color, i_source in source_sets:
fname = 'Data_PeptideOverlay/%s_motif.csv' % exp
for entry in datafile.read_csv(fname):
seqid = uniprot.parse_fasta_header(entry['Accessions'])[0]
if seqid not in protein:
protein[seqid] = default_protein()
protein[seqid]['description'] = entry['Names']
source = protein[seqid]['sources'][i_source]
source['color'] = color
source['peptides'].append(entry['sequence'])
seqids, fasta = uniprot.read_fasta('../db/uniprot_sprot.fasta')
for seqid in protein:
sequence = fasta[seqid]['sequence']
protein[seqid]['sequence'] = sequence
protein[seqid]['length'] = len(sequence)
for source in protein[seqid]['sources']:
for peptide in source['peptides']:
i = sequence.index(peptide)
j = i + len(peptide)
source['intervals'].append([i, j])
del source['peptides']
print 'write overlay%s.csv' % skip
make_csv(protein, 'overlay%s.csv' % skip)
import os
import uniprot
import pprint
# Clean up caches
os.system('rm *output* *cache*')
# Example 1 - reading a fasta file
seqids, fastas = uniprot.read_fasta('example.fasta')
pprint.pprint(seqids, indent=2)
# Example 2 - batch read identifier mappings with
# prespecified identifier types
seqids = """
NP_000508.1 NP_001018081.3
""".split()
pairs = uniprot.batch_uniprot_id_mapping_pairs(
'P_REFSEQ_AC', 'ACC', seqids)
pprint.pprint(pairs, indent=2)
# Example 3 - sequential identifier mapping to UniProt
# identifiers using robust but slow method
fname = 'KL-A1_pep15.csv'
peptides = []
proteins = {}
for row in csv.DictReader(open(fname)):
seqid = row['Protein']
if seqid not in proteins:
proteins[seqid] = []
proteins[seqid].append(row)
row['matches'] = []
peptides.append(row)
seqids, fasta = uniprot.read_fasta('use.fasta')
for seqid, protein in proteins.items():
bare_seqid = clean_seqid(seqid)
full_sequence = fasta[bare_seqid]['sequence']
for i_peptide1, peptide1 in enumerate(protein):
seq1 = peptide1['Sequence']
i1 = full_sequence.find(seq1)
peptide1['Start'] = i1 + 1
peptide1['End'] = i1 + len(seq1)
for peptide2 in protein:
if peptide1 == peptide2:
continue
seq2 = peptide2['Sequence']
i2 = full_sequence.find(seq2)
match = {
import os
import uniprot
import pprint
import sys
# Clean up caches
os.system('rm cache*')
# Example 1 - reading a fasta file
seqids, fastas = uniprot.read_fasta('example.fasta')
pprint.pprint(seqids, indent=2)
# Example 2 - map identifiers for RefSeq to Uniprot
seqids = "NP_000508.1 NP_001018081.3".split()
pairs = uniprot.batch_uniprot_id_mapping_pairs(
'P_REFSEQ_AC', 'ACC', seqids)
pprint.pprint(pairs, indent=2)
# Example 2 - get UniProt metadata
uniprot_seqids = [j for i,j in pairs]
uniprot_data = uniprot.batch_uniprot_metadata(
uniprot_seqids, 'cache')
pprint.pprint(uniprot_data, indent=2)
# Example 3 - parse for isoforms in metadata
text = open('cache/metadata.0.txt').read()
uniprot_data = uniprot.parse_isoforms(text)
pprint.pprint(uniprot_data)
# Example 4 - chaining commands to map seqids
seqids = "EFG_MYCA1 YP_885981.1 ENSG00000196176 Q91ZU6-8".split()
