upstream-extended/community/tools/biopython/converters/utils.py
1#!/usr/bin/env python
2
3from Bio import Seq
4from Bio.Alphabet import generic_dna
5
6
7def check_gff(gff_iterator):
8 """
9 Check GFF files before feeding to SeqIO to be sure they have sequences.
10 """
11 for rec in gff_iterator:
12 if isinstance(rec.seq, Seq.UnknownSeq):
13 print("Warning: FASTA sequence not found for '%s' in GFF file" % (rec.id))
14 rec.seq.alphabet = generic_dna
15 yield flatten_features(rec)
16
17
18def flatten_features(rec):
19 """
20 Make sub_features in an input rec flat for output.
21
22 GenBank does not handle nested features, so we want to make
23 everything top level.
24 """
25 out = []
26 for f in rec.features:
27 cur = [f]
28 while len(cur) > 0:
29 nextf = []
30 for curf in cur:
31 out.append(curf)
32 if len(curf.sub_features) > 0:
33 nextf.extend(curf.sub_features)
34 cur = nextf
35 rec.features = out
36 return rec
37
38
39def fix_ncbi_id(fasta_iter):
40 """
41 GenBank identifiers can only be 16 characters; try to shorten NCBI.
42 """
43 for rec in fasta_iter:
44 if len(rec.name) > 16 and rec.name.find("|") > 0:
45 new_id = [x for x in rec.name.split("|") if x][0]
46 print("Warning: shortening NCBI name %s to %s" % (rec.id, new_id))
47 rec.id = new_id
48 rec.name = new_id
49 yield rec