def split_consensus_fasta(fasta, ref_name, out):
elements = []
out_dir = out+"/split_fasta/"
mccutils.mkdir(out_dir)
fasta_records = SeqIO.parse(fasta,"fasta")
for record in fasta_records:
fasta_name = str(record.id)
elements.append(fasta_name)
special_chars = [";","&","(",")","|","*","?","[","]","~","{","}","<","!","^",'"',"'","\\","$","/"]
for char in special_chars:
fasta_name = fasta_name.replace(char,"_")
tmp_fasta = out_dir+ref_name+"_"+fasta_name+".fasta.tmp"
with open(tmp_fasta,"w") as outfa:
outfa.write(">"+str(record.id)+"\n")
outfa.write(str(record.seq)+"\n")
fasta_lines = fix_fasta.fix_fasta_lines(tmp_fasta, 80)
out_fasta = out_dir+ref_name+"_"+fasta_name+".fasta"
with open(out_fasta,"w") as outfa:
for line in fasta_lines:
outfa.write(line+"\n")
mccutils.remove(tmp_fasta)
return elements
def fix_fasta_lines(infasta, outfasta, length=80):
lines = fix_fasta.fix_fasta_lines(infasta, length)
with open(outfasta, "w") as fa:
for line in lines:
fa.write(line+"\n")
return outfasta
def main():
mccutils.log("processing", "making coverage fasta")
fastas = []
try:
length = 80
if snakemake.params.coverage_fasta == "None":
mccutils.run_command(["touch", snakemake.output.coverage_fasta])
else:
fasta3 = snakemake.params.coverage_fasta
fastas.append(fasta3)
lines = fix_fasta.fix_fasta_lines(fasta3, length)
write_fasta(lines, snakemake.output.coverage_fasta)
except Exception as e:
track = traceback.format_exc()
print(track, file=sys.stderr)
print(
"ERROR...failed to create coverage fasta, check the formatting of :",
snakemake.params.coverage_fasta,
file=sys.stderr)
mccutils.remove(snakemake.output[0])
mccutils.remove(snakemake.output[1])
mccutils.remove(snakemake.output[2])
sys.exit(1)
mccutils.log("processing", "coverage fasta created")
def repeat_mask(reference, te_fasta, chromosomes, procs, run_id, log, out):
try:
outdir = out + "/tmp/repeatmasker_" + run_id
mccutils.mkdir(outdir)
os.chdir(outdir)
command = [
"RepeatMasker", "-pa",
str(procs), "-lib", te_fasta, "-dir", outdir, "-s", "-gff",
"-nolow", "-no_is", reference
]
mccutils.run_command(command, log=log)
os.chdir(out)
# RepeatMasker appears to override the custom database names during the ProcessRepeats
# this step changes them back, more rules may be needed for other reference genomes
ref_name = os.path.basename(reference)
repeatmasker_gff = outdir + "/" + ref_name + ".out.gff"
formatted_ref_tes = out + "/tmp/" + run_id + "tmpreferenceTEs.gff"
with open(repeatmasker_gff, "r") as rmgff:
with open(formatted_ref_tes, "w") as outgff:
for line in rmgff:
if "#" not in line:
line = line.replace("McClintock-int", "McClintock")
line = line.replace("POGON1", "pogo")
split_line = line.split("\t")
feats = split_line[8]
if split_line[0] in chromosomes:
te = feats.split(" ")[1]
te = te.replace('"', '').split(":")[1]
feats = ";".join(
["ID=" + te, "Name=" + te, "Alias=" + te])
split_line[2] = te
split_line[8] = feats
line = "\t".join(split_line)
outgff.write(line + '\n')
masked_fasta = outdir + "/" + ref_name + ".masked"
fasta_lines = fix_fasta.fix_fasta_lines(masked_fasta, 80)
mccutils.check_file_exists(formatted_ref_tes)
except Exception as e:
track = traceback.format_exc()
print(track, file=sys.stderr)
print("ERROR...Failed to run repeatmasker on: ",
reference,
"with lib:",
te_fasta,
"check file formatting...exiting...",
file=sys.stderr)
sys.exit(1)
return formatted_ref_tes