def tbprofiler(fq1,fq2,uniq_id,db,storage_dir,platform):
conf = get_conf_dict(sys.base_prefix+"/share/tbprofiler/tbdb")
drug_order = ["isoniazid","rifampicin","ethambutol","pyrazinamide","streptomycin","ethionamide","fluoroquinolones","amikacin","capreomycin","kanamycin"]
if fq1 and fq2:
fastq_obj = pp.fastq(fq1,fq2)
elif fq1 and fq2==None:
fastq_obj = pp.fastq(fq1)
files_prefix = storage_dir+"/"+uniq_id
bam_obj = fastq_obj.map_to_ref(
ref_file=conf["ref"], prefix=files_prefix,sample_name=uniq_id,
aligner="bwa", platform=platform, threads=4
)
bam_file = bam_obj.bam_file
results = pp.bam_profiler(
conf=conf, bam_file=bam_file, prefix=files_prefix, platform=platform,
caller="bcftools", threads=4, no_flagstat=False,
run_delly = True
)
results = tbp.reformat(results, conf, reporting_af=0.1)
results["id"] = uniq_id
results["tbprofiler_version"] = tbp._VERSION
results["pipeline"] = {"mapper":"bcftools","variant_caller":"bcftools"}
results = tbp.get_summary(results,conf,drug_order=drug_order)
outfile = "%s.results.json" % (storage_dir+"/"+uniq_id)
json.dump(results,open(outfile,"w"))
conn = sqlite3.connect(db)
c = conn.cursor()
c.execute("UPDATE results SET result = ?, lineage = ?, drtype = ?, status = 'completed' where id = ?", (open(outfile).readline(),results["sublin"],results["drtype"],uniq_id,))
c.execute("UPDATE full_results SET main_lineage = ?, sub_lineage = ?, DR_type = ?, MDR = ?, XDR = ?",(results["main_lin"],results["sublin"],results["drtype"],results["MDR"],results["XDR"]))
for d in results["drug_table"]:
c.execute("UPDATE full_results SET %s = ? where id = ?" % d["Drug"].lower().replace("-","_"), (d["Mutations"],uniq_id,))
conn.commit()
pp.run_cmd("rm %s/%s*" % (storage_dir,uniq_id))
return True
def main_profile(args):
#### Setup conf dictionary ###
if args.db == "tbdb" and not args.external_db and pp.nofile(
sys.base_prefix + "/share/tbprofiler/tbdb.fasta"):
pp.log(
"Can't find the tbdb file at %s. Please run 'tb-profiler update_tbdb' to load the default library or specify another using the '--external_db' flag"
% sys.base_prefix,
ext=True)
if args.external_db:
conf = get_conf_dict(args.external_db)
else:
conf = get_conf_dict(sys.base_prefix +
"/share/tbprofiler/%s" % args.db)
### Create folders for results if they don't exist ###
if pp.nofolder(args.dir):
os.mkdir(args.dir)
for x in ["bam", "vcf", "results"]:
if pp.nofolder(args.dir + "/" + x):
os.mkdir(args.dir + "/" + x)
### Set up platform dependant parameters ###
if args.platform == "nanopore":
args.mapper = "minimap2"
args.caller = "bcftools"
args.no_trim = True
run_delly = False
else:
if args.no_delly:
run_delly = False
else:
run_delly = True
### Setup prefix for files ###
files_prefix = args.dir + "/" + args.prefix
### Create bam file if fastq has been supplied ###
if args.bam == None:
if args.read1 and args.read2 and args.no_trim:
# Paired + no trimming
fastq_obj = pp.fastq(args.read1, args.read2)
elif args.read1 and args.read2 and not args.no_trim:
# Paired + trimming
untrimmed_fastq_obj = pp.fastq(args.read1, args.read2)
fastq_obj = untrimmed_fastq_obj.trim(files_prefix,
threads=args.threads)
elif args.read1 and not args.read2 and args.no_trim:
# Unpaired + trimming
fastq_obj = pp.fastq(args.read1, args.read2)
elif args.read1 and not args.read2 and not args.no_trim:
# Unpaired + trimming
untrimmed_fastq_obj = pp.fastq(args.read1)
fastq_obj = untrimmed_fastq_obj.trim(files_prefix,
threads=args.threads)
else:
exit("\nPlease provide a bam file or a fastq file(s)...Exiting!\n")
bam_obj = fastq_obj.map_to_ref(ref_file=conf["ref"],
prefix=files_prefix,
sample_name=args.prefix,
aligner=args.mapper,
platform=args.platform,
threads=args.threads)
bam_file = bam_obj.bam_file
else:
bam_file = args.bam
print(args.delly_bcf_file)
run_coverage = False if args.no_coverage else True
### Run profiling module from pathogen-profiler ###
results = pp.bam_profiler(
conf=conf,
bam_file=bam_file,
prefix=files_prefix,
platform=args.platform,
caller=args.caller,
threads=args.threads,
no_flagstat=args.no_flagstat,
run_delly=run_delly,
calling_params=args.calling_params,
coverage_fraction_threshold=args.coverage_fraction_threshold,
missing_cov_threshold=args.missing_cov_threshold,
delly_bcf_file=args.delly_bcf_file)
json.dump(results, open(args.prefix + ".tmp_results.json", "w"))
### Reformat the results to TB-Profiler style ###
results = tbp.reformat(results, conf, reporting_af=args.reporting_af)
results["id"] = args.prefix
results["tbprofiler_version"] = tbp._VERSION
results["pipeline"] = {
"mapper": args.mapper if not args.bam else "N/A",
"variant_caller": args.caller
}
json_output = args.dir + "/results/" + args.prefix + ".results.json"
tex_output = args.dir + "/results/" + args.prefix + ".results.tex"
text_output = args.dir + "/results/" + args.prefix + ".results.txt"
csv_output = args.dir + "/results/" + args.prefix + ".results.csv"
json.dump(results, open(json_output, "w"))
extra_columns = [x.lower() for x in args.add_columns.split(",")
] if args.add_columns else []
if args.pdf:
tbp.write_tex(results, conf, tex_output, extra_columns)
pp.run_cmd("pdflatex %s" % tex_output, verbose=1)
pp.rm_files([
tex_output, args.dir + "/" + args.prefix + ".results.aux",
args.dir + "/" + args.prefix + ".results.log"
])
if args.txt:
tbp.write_text(results,
conf,
text_output,
extra_columns,
reporting_af=args.reporting_af)
if args.csv:
tbp.write_csv(results, conf, csv_output, extra_columns)
### Move files to respective directories ###
if not args.bam:
pp.run_cmd("mv %(dir)s/%(prefix)s.bam* %(dir)s/bam/" % vars(args))
if not args.no_trim:
pp.run_cmd("rm -f %s" % " ".join(fastq_obj.files))
pp.run_cmd("mv -f %(dir)s/%(prefix)s*.vcf.gz* %(dir)s/vcf/" % vars(args))
if run_delly and results["delly"] == "success" and not args.delly_bcf_file:
pp.run_cmd("mv -f %(dir)s/%(prefix)s.delly.bcf* %(dir)s/vcf/" %
vars(args))
### Add meta data to results
if args.meta:
for row in csv.DictReader(open(args.meta)):
if row["id"] == results["id"]:
for col in row:
results["meta_" + col] = row[col]
pp.log("Profiling finished sucessfully!")
def main_profile(args):
if pp.nofolder(args.dir):
os.mkdir(args.dir)
conf = get_conf_dict(sys.base_prefix + "/share/covidprofiler/%s" % args.db)
### Setup prefix for files ###
files_prefix = args.dir + "/" + args.prefix
if args.fasta:
if args.read1 or args.read2:
sys.stderr.write(
"Please use --fasta or --read1/2 but not both... Exiting!\n")
quit()
fasta_obj = pp.fasta(args.fasta)
wg_vcf_obj = pp.vcf(
fasta_obj.get_ref_variants(conf["ref"],
prefix=args.prefix,
file_prefix=files_prefix))
else:
if not args.read1:
sys.stderr.write(
"Please provide assembly using --fasta or at least one read file using --read1... Exiting!\n"
)
quit()
### Create bam file if fastq has been supplied ###
if args.read1 and args.read2 and args.no_trim:
# Paired + no trimming
fastq_obj = pp.fastq(args.read1, args.read2)
elif args.read1 and args.read2 and not args.no_trim:
# Paired + trimming
untrimmed_fastq_obj = pp.fastq(args.read1, args.read2)
fastq_obj = untrimmed_fastq_obj.trim(files_prefix,
threads=args.threads)
elif args.read1 and not args.read2 and args.no_trim:
# Unpaired + trimming
fastq_obj = pp.fastq(args.read1, args.read2)
elif args.read1 and not args.read2 and not args.no_trim:
# Unpaired + trimming
untrimmed_fastq_obj = pp.fastq(args.read1)
fastq_obj = untrimmed_fastq_obj.trim(files_prefix,
threads=args.threads)
bam_obj = fastq_obj.map_to_ref(ref_file=conf["ref"],
prefix=files_prefix,
sample_name=args.prefix,
aligner=args.mapper,
platform=args.platform,
threads=args.threads)
wg_vcf_obj = bam_obj.call_variants(conf["ref"],
args.caller,
remove_missing=True)
cp.vcf2consensus(bam_obj.bam_file, wg_vcf_obj.filename, conf["ref"],
wg_vcf_obj.samples[0],
wg_vcf_obj.prefix + ".consensus.fasta")
if not args.no_trim:
pp.run_cmd("rm -f %s" % " ".join(fastq_obj.files))
refseq = pp.fasta(conf["ref"]).fa_dict
refseqname = list(refseq.keys())[0]
results = {}
barcode_mutations = wg_vcf_obj.get_bed_gt(conf["barcode"], conf["ref"])
barcode = pp.barcode(barcode_mutations, conf["barcode"])
clade = ";".join(sorted([d["annotation"] for d in barcode]))
sys.stdout.write("%s\t%s\n" % (args.prefix, clade))
results["clade"] = clade
variant_data = cp.get_variant_data(wg_vcf_obj.filename, conf["ref"],
conf["gff"], conf["proteins"])
results["variants"] = variant_data
json.dump(results, open("%s.results.json" % files_prefix, "w"))