def process(args):
out_script = "%s.run.sh" % args.prefix
O = open(out_script, "w")
ps.filecheck(args.ref)
samples = []
args.sample_file = "%s.samples.txt" % args.prefix
for row in csv.DictReader(open(args.csv)):
args.id = row["ID"]
samples.append(row["ID"])
args.r1 = "%s/%s" % (args.fastq_dir, row["R1"])
ps.filecheck(args.r1)
#params["centrifuge"] = "--centrifuge %s" % args.centrifuge if args.centrifuge else ""
args.r2 = "%s/%s" % (args.fastq_dir, row["R2"])
ps.filecheck(args.r2)
O.write(
"illumina_pipeline.py %(ref)s %(r1)s %(r2)s %(id)s -t %(threads)s -m %(mapper)s \n"
% vars(args))
O.write(
"tb-profiler profile -p %(id)s -a %(id)s.bam -t %(threads)s\n" %
vars(args))
O.write("merge_vcfs.py %(sample_file)s %(ref)s %(prefix)s\n" % vars(args))
args.snps_aln_file = "%s.snps.fa" % args.prefix
O.write(
"raxml-ng --msa %(snps_aln_file)s --search --model GTR+G --threads `raxml-ng --msa %(snps_aln_file)s --parse --model GTR+G | grep MPI | awk '{print $9}'`\n"
% vars(args))
O.write("tb-profiler collate\n")
O.close()
open(args.sample_file, "w").write("\n".join(samples))
if not args.dry:
ps.run_cmd("bash %s" % out_script)
def reheader(bamfile,oldname,newname,threads):
tmpfile = "%s.header" % bamfile
cmd = "samtools view %s -H | sed 's/%s/%s/g' > %s" % (bamfile,oldname,newname,tmpfile)
ps.run_cmd(cmd)
newbamfile = "%s.reheader.bam" % (bamfile.replace(".bam",""))
cmd = "samtools reheader %s %s | samtools view -@ %s -b > %s" % (tmpfile,bamfile,threads,newbamfile)
ps.run_cmd(cmd)
def main(args):
bcf = ps.bcf(args.bcf)
bcf.get_mean_genotype()
bcf.get_genesum()
geno_file = bcf.prefix+".geno"
genesum_file = bcf.prefix+".genesum"
meta = {}
for s in bcf.samples:
meta[s] = {}
for row in csv.DictReader(open(args.csv)):
for pheno in row.keys():
if pheno=="id": continue
if row['id'] not in meta: continue
meta[row["id"]][pheno] = row[pheno]
phenos = [x.rstrip() for x in open(args.phenos).readlines()]
cmd_file = ps.get_random_file()
X = open(cmd_file,"w")
for pheno in phenos:
pheno_file = "%s.pheno" % pheno
if pheno not in row:
ps.log("%s not in CSV file"%pheno,True)
P = open(pheno_file,"w")
P.write("\n".join([meta[s][pheno] if pheno in meta[s] else "NA" for s in bcf.samples]))
P.close()
X.write("gemma -p %s -g %s -gk 1 -o %s -maf 0.00005 -miss 0.99 && gemma -lmm 1 -p %s -g %s -k output/%s.cXX.txt -o %s -maf 0.00005 -miss 0.99 && gemma -lmm 1 -p %s -g %s -k output/%s.cXX.txt -o %s.genesum -notsnp\n" % (pheno_file,geno_file,pheno,pheno_file,geno_file,pheno,pheno,pheno_file,genesum_file,pheno,pheno))
X.close()
if args.preprocess:
ps.log("Preprocessing finished\n", True)
else:
ps.run_cmd("cat %s | parallel -j %s" % (cmd_file,args.threads))
def main(args):
bcf_file = args.bcf
ref_file = args.ref
bcf = ps.bcf(bcf_file)
bcf.generate_consensus(ref_file,
threads=args.threads,
no_chrom=args.combine)
if args.combine:
files = " ".join(
["%s.%s.fasta" % (bcf.prefix, s) for s in bcf.samples])
ps.run_cmd("cat %s > %s.genome.fasta" % (files, bcf.prefix))
ps.run_cmd("rm %s" % (files))
def process(args):
out_script = "%s.run.sh" % args.prefix
O = open(out_script,"w")
ps.filecheck(args.ref)
samples = []
args.sample_file = "%s.samples.txt" % args.prefix
for row in csv.DictReader(open(args.csv)):
args.id = row["ID"]
samples.append(row["ID"])
args.r1 = "%s/%s" % (args.fastq_dir,row["R1"])
ps.filecheck(args.r1)
#params["centrifuge"] = "--centrifuge %s" % args.centrifuge if args.centrifuge else ""
args.r2 = "%s/%s" % (args.fastq_dir,row["R2"])
ps.filecheck(args.r2)
O.write("illumina_pipeline.py %(ref)s %(r1)s %(r2)s %(id)s -t %(threads)s -m %(mapper)s \n" % vars(args))
O.close()
open(args.sample_file,"w").write("\n".join(samples))
if not args.dry:
ps.run_cmd("bash %s" % out_script)
def process(args):
out_script = "%s.run.sh" % args.prefix
O = open(out_script, "w")
ps.filecheck(args.ref)
samples = []
args.sample_file = "%s.samples.txt" % args.prefix
for row in csv.DictReader(open(args.csv)):
args.id = row["ID"]
samples.append(row["ID"])
args.r1 = "%s/%s" % (args.fastq_dir, row["R1"])
ps.filecheck(args.r1)
#params["centrifuge"] = "--centrifuge %s" % args.centrifuge if args.centrifuge else ""
args.r2 = "%s/%s" % (args.fastq_dir, row["R2"])
ps.filecheck(args.r2)
O.write(
"illumina_pipeline.py %(ref)s %(r1)s %(r2)s %(id)s -t %(threads)s -m %(mapper)s \n"
% vars(args))
O.close()
open(args.sample_file, "w").write("\n".join(samples))
if not args.dry:
ps.run_cmd("bash %s" % out_script)
def main(args):
bcf = ps.bcf(args.bcf)
bcf.get_mean_genotype()
bcf.get_genesum()
geno_file = bcf.prefix + ".geno"
genesum_file = bcf.prefix + ".genesum"
meta = {}
for s in bcf.samples:
meta[s] = {}
for row in csv.DictReader(open(args.csv)):
for pheno in row.keys():
if pheno == "id": continue
if row['id'] not in meta: continue
meta[row["id"]][pheno] = row[pheno]
phenos = [x.rstrip() for x in open(args.phenos).readlines()]
cmd_file = ps.get_random_file()
X = open(cmd_file, "w")
for pheno in phenos:
pheno_file = "%s.pheno" % pheno
if pheno not in row:
ps.log("%s not in CSV file" % pheno, True)
P = open(pheno_file, "w")
P.write("\n".join([
meta[s][pheno] if pheno in meta[s] else "NA" for s in bcf.samples
]))
P.close()
X.write(
"gemma -p %s -g %s -gk 1 -o %s -maf 0.00005 -miss 0.99 && gemma -lmm 1 -p %s -g %s -k output/%s.cXX.txt -o %s -maf 0.00005 -miss 0.99 && gemma -lmm 1 -p %s -g %s -k output/%s.cXX.txt -o %s.genesum -notsnp\n"
% (pheno_file, geno_file, pheno, pheno_file, geno_file, pheno,
pheno, pheno_file, genesum_file, pheno, pheno))
X.close()
if args.preprocess:
ps.log("Preprocessing finished\n", True)
else:
ps.run_cmd("cat %s | parallel -j %s" % (cmd_file, args.threads))
def merge_sample(sample, cram=False, reference=None, threads=20):
cram_flag = "-CT %s" % reference if cram else "-b"
extension = ".cram" if cram else ".bam"
samples = sample.split("_")
for x in samples:
ps.filecheck(x + extension)
bams = " ".join([x + extension for x in samples])
first_bam = samples[0] + extension
temp_bam = sample + ".tmp.bam"
final_bam = sample + extension
header = sample + ".header"
cmd = "samtools view %s -H | sed 's/%s/%s/g' > %s" % (
first_bam, samples[0], sample, header)
ps.run_cmd(cmd)
cmd = "samtools merge -@ %s %s %s" % (threads, temp_bam, bams)
ps.run_cmd(cmd)
cmd = "samtools reheader %s %s | samtools view -@ %s %s > %s" % (
header, temp_bam, threads, cram_flag, final_bam)
ps.run_cmd(cmd)
cmd = "rm %s %s" % (header, temp_bam)
ps.run_cmd(cmd)
#! /usr/bin/env python import sys import pathogenseq as ps ref_file = sys.argv[1] query_file = sys.argv[2] prefix = sys.argv[3] ps.mauve_call_variants(ref_file,query_file,prefix) cmd = "bgzip -f %s.vcf" % prefix ps.run_cmd(cmd)
#! /usr/bin/env python
import sys
import pathogenseq as ps
import json
args = {}
args["r1"] = sys.argv[1]
args["r2"] = sys.argv[2]
args["prefix"] = sys.argv[3]
args["centrifuge_db"] = sys.argv[4]
args["fasta_db"] = sys.argv[5]
args["threads"] = sys.argv[6]
args["fq_report"] = ps.get_random_file()
args["log"] = ps.get_random_file()
cmd = "centrifuge -x %(centrifuge_db)s -1 %(r1)s -2 %(r2)s -S %(log)s --report-file %(fq_report)s -p %(threads)s" % args
ps.run_cmd(cmd)
best_ref = ""
best_score = 0
best_species = ""
best_species_score = 0
for l in open(args["fq_report"]):
row = l.rstrip().split("\t")
if row[4] == "numReads": continue
if row[2] == "leaf" and int(row[4]) > best_score:
best_ref = row[0]
best_score = int(row[4])
elif row[2] == "species" and int(row[4]) > best_species_score:
best_species = row[0]
best_species_score = int(row[4])