Mercurial > repos > iuc > transit_resampling

comparison gff_to_prot.py @ 2:173bf0f6ca1e draft

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-iuc/tree/master/tools/transit/ commit 9ffba16c072c7dafe4ef0b70cd2ef166fbe903d0-dirty
author iuc
date Thu, 28 Feb 2019 09:04:07 -0500
parents
children 3fcb70c1ca78
comparison
equal deleted inserted replaced
1:1d1ca6b30fbb 2:173bf0f6ca1e
1 #!/usr/bin/env python
2 import sys
3 import os
4 import csv
5
6 def get_description(line, parent):
7 cols = line.split('\t')
8 labels = {}
9 for pair in cols[8].split(";"):
10 k, v = pair.split('=')
11 labels[k] = v
12
13 if (cols[2]) == "CDS" and labels["Parent"] == parent:
14 return labels.get("Note", '-')
15 return '-'
16
17
18 def convert_to_prot_table(fname, output_name):
19 gff_file = open(fname)
20 output_file = open(output_name, 'w')
21 writer = csv.writer(output_file, delimiter='\t')
22 lines = gff_file.readlines()
23 gff_file.close()
24 for i, line in enumerate(lines):
25 lie = line.strip()
26 if line.startswith('#'): continue
27 cols = line.split('\t')
28 if (len(cols) < 9):
29 print("Ignoring invalid row with entries: {0}".format(cols))
30 elif (cols[2]) == "region": continue
31 elif (cols[2]) == "CDS": continue
32 elif (cols[2]) == "gene":
33 start = int(cols[3])
34 end = int(cols[4])
35 strand = cols[6].strip()
36 labels = {}
37 diff = int(abs(end - start)/3) ## What is this called?
38 for pair in cols[8].split(";"):
39 k, v = pair.split('=')
40 labels[k.strip()] = v.strip()
41
42 Rv = labels["locus_tag"].strip() # error out if not found
43 gene = labels.get('Name', '')
44 desc = get_description(lines[i + 1], labels.get("ID", "")) if (i + 1) < len(lines) else '-'
45 vals = [desc, start, end, strand, diff, '-', '-', gene, Rv, '-']
46 writer.writerow(vals)
47 output_file.close()
48
49 if __name__ == "__main__":
50 usage_string = "Usage: python gff-prot-converter.py <gff filename> <output filename>"
51
52 if len(sys.argv) < 3:
53 print(usage_string)
54 sys.exit(0)
55 file_name = sys.argv[1]
56 if not os.path.exists(file_name):
57 print("File not found. Exiting...")
58 print(usage_string)
59 sys.exit(0)
60 convert_to_prot_table(file_name, sys.argv[2])