Mercurial > repos > davidvanzessen > shm_csr

comparison igm_naive_mutations.py @ 92:cf8ad181628f draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload commit 36be3b053802693392f935e6619ba3f2b1704e3c
author rhpvorderman
date Mon, 12 Dec 2022 12:32:44 +0000
parents
children
comparison
equal deleted inserted replaced
91:f387cc1580c6 92:cf8ad181628f
1 #!/usr/bin/env python3
2
3 """
4 Find naive mutations (< 2% mutated) for IGM genes
5 """
6
7 import argparse
8 import contextlib
9
10
11 def find_naive_mutations(mutation_file, naive_file, naive_memory_file,
12 percentage_cutoff=0.02):
13 # A compound with statement throws a syntax error with the included python
14 # 3.7.1 in the container, so use an exit stack instead.
15 with contextlib.ExitStack() as stack:
16 mutations = stack.enter_context(open(mutation_file, "rt"))
17 naive = stack.enter_context(open(naive_file, "wt"))
18 naive_memory = stack.enter_context(open(naive_memory_file, "wt"))
19 header = next(mutations)
20 naive.write(header)
21 naive_memory.write(header)
22 for line in mutations:
23 sequence_id, best_match, mutation_no, region_length, _ = \
24 line.strip('\n').split('\t')
25 if best_match != "IGM":
26 continue
27 mutation_no = int(mutation_no)
28 region_length = int(region_length)
29 if (mutation_no / region_length) < percentage_cutoff:
30 naive.write(line)
31 else:
32 naive_memory.write(line)
33
34
35 def main():
36 parser = argparse.ArgumentParser()
37 parser.add_argument("mutation_file", help="scatter.txt")
38 parser.add_argument("naive_file")
39 parser.add_argument("naive_memory_file")
40 args = parser.parse_args()
41 find_naive_mutations(args.mutation_file, args.naive_file,
42 args.naive_memory_file)
43
44
45 if __name__ == "__main__":
46 main()