def main_load_library(args):
lib_prefix = args.prefix.split("/")[-1]
files = {
"gff": ".gff",
"ref": ".fasta",
"barcode": ".barcode.bed",
"version": ".version.json"
}
if pp.nofolder(sys.base_prefix + "/share/covidprofiler"):
pp.run_cmd("mkdir %s " % (sys.base_prefix + "/share/covidprofiler/"))
for key in files:
new_file_location = sys.base_prefix + "/share/covidprofiler/" + lib_prefix + files[
key]
pp.run_cmd("cp %s %s" % (args.prefix + files[key], new_file_location))
pp.run_cmd("samtools faidx %s" % sys.base_prefix +
"/share/covidprofiler/" + lib_prefix + ".fasta")
pp.run_cmd("bwa index %s" % sys.base_prefix + "/share/covidprofiler/" +
lib_prefix + ".fasta")
if os.path.isfile("%s" % sys.base_prefix + "/share/covidprofiler/" +
lib_prefix + ".dict"):
pp.run_cmd("rm %s" % sys.base_prefix + "/share/covidprofiler/" +
lib_prefix + ".dict")
pp.run_cmd("gatk CreateSequenceDictionary -R %s" % sys.base_prefix +
"/share/covidprofiler/" + lib_prefix + ".fasta")
pp.log("Sucessfully imported library")
def main_load_library(args):
lib_prefix = args.prefix.split("/")[-1]
files = {
"gff": ".gff",
"ref": ".fasta",
"barcode": ".barcode.bed",
"version": ".version.json",
"proteins": ".proteins.csv",
"non_coding_bed": ".non_coding.bed"
}
if pp.nofolder(sys.base_prefix + "/share/covidprofiler"):
pp.run_cmd("mkdir %s " % (sys.base_prefix + "/share/covidprofiler/"))
pp.run_cmd("cp %s %s" % (args.msa, "%s/share/covidprofiler/%s.msa.fa" %
(sys.base_prefix, lib_prefix)))
pp.run_cmd("cp %s %s" %
(args.meta, "%s/share/covidprofiler/%s.msa.meta.csv" %
(sys.base_prefix, lib_prefix)))
for key in files:
new_file_location = sys.base_prefix + "/share/covidprofiler/" + lib_prefix + files[
key]
pp.run_cmd("cp %s %s" % (args.prefix + files[key], new_file_location))
pp.run_cmd("samtools faidx %s" % sys.base_prefix +
"/share/covidprofiler/" + lib_prefix + ".fasta")
pp.run_cmd("bwa index %s" % sys.base_prefix + "/share/covidprofiler/" +
lib_prefix + ".fasta")
if os.path.isfile("%s" % sys.base_prefix + "/share/covidprofiler/" +
lib_prefix + ".dict"):
pp.run_cmd("rm %s" % sys.base_prefix + "/share/covidprofiler/" +
lib_prefix + ".dict")
pp.log("Sucessfully imported library")
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"))
def main(args):
if pp.nofolder(args.out_dir):
pp.run_cmd("mkdir %s" % args.out_dir)
conf = {
"ref": args.ref,
"gff": args.gff,
"bed": args.bed,
"ann": args.ann,
}
if args.conf:
conf = json.load(open(args.conf))
for x in ["ref", "gff", "bed", "ann"]:
if conf[x] == None:
pp.log("%s variable is not defined" % x, True)
bam_obj = pp.bam(args.bam,
args.prefix,
conf["ref"],
platform=args.platform)
bcf_obj = bam_obj.call_variants(
prefix=args.prefix + ".targets",
call_method=args.call_method,
gff_file=conf["gff"],
bed_file=conf["bed"],
mixed_as_missing=False if args.platform == "Illumina" else True,
threads=args.threads,
min_dp=args.min_depth,
af=args.af,
caller=args.caller)
csq = bcf_obj.load_csq(ann_file=conf["ann"])
variants = []
chr2gene_pos = {}
for l in open(conf["ann"]):
row = l.rstrip().split()
chr2gene_pos[int(row[1])] = int(row[3])
for var in list(csq.values())[0]:
var["_internal_change"] = var["change"]
var["change"] = pp.reformat_mutations(var["change"], var["type"],
var["gene_id"], chr2gene_pos)
variants.append(var)
if not args.no_delly:
delly_bcf = bam_obj.run_delly()
deletions = delly_bcf.overlap_bed(conf["bed"])
for deletion in deletions:
tmp_change = pp.reformat_mutations(
"%(chr)s_%(start)s_%(end)s" % deletion, var["type"],
var["gene_id"], chr2gene_pos)
tmp = {
"genome_pos": deletion["start"],
"gene_id": deletion["region"],
"chr": deletion["chr"],
"freq": 1,
"type": "large_deletion",
"change": tmp_change
}
variants.append(tmp)
json.dump(variants,
open("%s/%s.pp-results.json" % (args.out_dir, args.prefix), "w"))
for x in [
".targets.bcf", ".targets.csq.bcf", ".targets.csq.bcf.csi",
".targets.delly.bcf", ".targets.delly.bcf.csi",
".targets.del_pos.bed", ".targets.gvcf.gz", ".targets.gvcf.gz.csi",
".targets.missing.bcf"
]:
if args.no_delly and "delly" in x: continue
pp.run_cmd("rm %s%s" % (args.prefix, x))