Mercurial > repos > devteam > fasta_compute_length

comparison fasta_compute_length.py @ 6:5cbde03c1103 draft default tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/galaxyproject/tools-devteam/tree/master/tools/fasta_compute_length commit 9f0b8eb9fd7cf9e2513e7c822767153dbd4bc3b2
author devteam
date Thu, 03 Oct 2024 12:11:55 +0000
parents e12f68d2cc4e
children
comparison
equal deleted inserted replaced
5:7d37cfda8e00 6:5cbde03c1103
1 #!/usr/bin/env python 1 #!/usr/bin/env python
2 """ 2 """
3 Uses fasta_to_len converter code. 3 Input: fasta, int
4 Output: tabular
5 Return titles with lengths of corresponding seq
4 """ 6 """
5 7
6 import sys 8 import sys
7 from utils.fasta_to_len import compute_fasta_length
8 9
9 10
10 compute_fasta_length(sys.argv[1], sys.argv[2], sys.argv[3], sys.argv[4] == 'id_only') 11 def compute_fasta_length(fasta_file, out_file, keep_first_char, keep_first_word=False):
12 keep_first_char = int(keep_first_char)
13 fasta_title = ''
14 seq_len = 0
15
16 # number of char to keep in the title
17 if keep_first_char == 0:
18 keep_first_char = None
19 else:
20 keep_first_char += 1
21
22 first_entry = True
23 with open(fasta_file) as in_fh, open(out_file, 'w') as out_fh:
24 for line in in_fh:
25 line = line.strip()
26 if not line or line.startswith('#'):
27 continue
28 if line[0] == '>':
29 if first_entry is False:
30 if keep_first_word:
31 fasta_title = fasta_title.split()[0]
32 out_fh.write("%s\t%d\n" % (fasta_title[1:keep_first_char], seq_len))
33 else:
34 first_entry = False
35 fasta_title = line
36 seq_len = 0
37 else:
38 seq_len += len(line)
39
40 # last fasta-entry
41 if keep_first_word:
42 fasta_title = fasta_title.split()[0]
43 out_fh.write("%s\t%d\n" % (fasta_title[1:keep_first_char], seq_len))
44
45
46 if __name__ == "__main__":
47 compute_fasta_length(sys.argv[1], sys.argv[2], sys.argv[3], sys.argv[4] == 'id_only')