Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions .github/workflows/main.yml
Original file line number Diff line number Diff line change
Expand Up @@ -2,9 +2,9 @@ name: Tests

on:
push:
branches: [main, dev]
branches: [main]
pull_request:
branches: [main, dev]
branches: [main]

jobs:
GenerateMatrix:
Expand Down
31 changes: 25 additions & 6 deletions get_genome/wrapper.py
Original file line number Diff line number Diff line change
Expand Up @@ -37,12 +37,12 @@
messages = []


def check_fasta(fasta, messages=[]):
with open(fasta, "r") as fasta_file:
def check_fasta(input_fasta, messages=[]):
with open(input_fasta, "r") as fasta_file:
fasta = fasta_file.read()
n_items = fasta.count(">")
if n_items:
messages += [f"Supplied fasta file '{fasta}' was found"]
messages += [f"Supplied fasta file '{input_fasta}' was found"]
messages += [f"Supplied fasta file contains {n_items} items"]
else:
raise ValueError(
Expand Down Expand Up @@ -71,21 +71,36 @@ def check_gff(input_gff, messages=[]):
gff_source_type += list(i.items())
limits = dict(gff_source_type=gff_source_type)
for rec in GFF.parse(gff_file, limit_info=limits):
unique_names = set()
for recfeat in rec.features:
if recfeat.type == "inferred_parent":
continue
rec_keys = recfeat.qualifiers.keys()
if not "Name" in rec_keys:
if "locus_tag" in rec_keys:
messages += [
f"Added missing 'Name' from locus_tag: {recfeat.qualifiers["locus_tag"][0]}"
]
recfeat.qualifiers["Name"] = recfeat.qualifiers["locus_tag"]
else:
raise ValueError(
"required fields 'Name','locus_tag' missing in *.gff file"
)
else:
if "locus_tag" in rec_keys:
recfeat.qualifiers["trivial_name"] = recfeat.qualifiers["Name"]
recfeat.qualifiers["Name"] = recfeat.qualifiers["locus_tag"]
if recfeat.qualifiers["Name"][0] in unique_names:
if "locus_tag" in rec_keys:
messages += [
f"Added locus_tag suffix to the duplicated 'Name': {recfeat.qualifiers["Name"][0]}"
]
recfeat.qualifiers["Name"] = (
f"{recfeat.qualifiers["Name"][0]}_{recfeat.qualifiers["locus_tag"][0]}"
)
else:
raise ValueError(
f"found duplicated 'Name' tag in *.gff: {recfeat.qualifiers["Name"][0]}"
)
else:
unique_names.add(recfeat.qualifiers["Name"][0])
if not "ID" in rec_keys:
if "locus_tag" in rec_keys:
recfeat.qualifiers["ID"] = recfeat.qualifiers["locus_tag"]
Expand Down Expand Up @@ -161,6 +176,8 @@ def check_gff(input_gff, messages=[]):
gff, messages = check_gff(output_gff, messages)
with open(output_gff, "w") as gff_out:
GFF.write(gff, gff_out)
with open(str(snakemake.log), "a") as log_out:
log_out.write("\n".join(messages) + "\n")

elif db.lower() == "manual":
if not path.exists(fasta):
Expand All @@ -176,6 +193,8 @@ def check_gff(input_gff, messages=[]):
fasta_out.write(fasta)
with open(output_gff, "w") as gff_out:
GFF.write(gff, gff_out)
with open(str(snakemake.log), "a") as log_out:
log_out.write("\n".join(messages) + "\n")
# index FASTA file
shell("samtools faidx {output_fasta} {log}")
else:
Expand Down