def prepare_genes(self):
curl_ensembl_gtf = Curl()
curl_ensembl_gtf.remote = self.ensembl_gtf_remote
curl_ensembl_gtf.output = "{}/genes/{}".format(
self.outdir, os.path.basename(self.ensembl_gtf_remote))
curl_ensembl_gtf.jobname = "curl-ensembl-gtf"
curl_ensembl_gtf.is_intermediate = True
self.add(curl_ensembl_gtf)
gunzip_ensembl_gtf = Gunzip()
gunzip_ensembl_gtf.input = curl_ensembl_gtf.output
gunzip_ensembl_gtf.output = stripsuffix(curl_ensembl_gtf.output, ".gz")
gunzip_ensembl_gtf.is_intermediate = True
self.add(gunzip_ensembl_gtf)
filt_ensembl_gtf_chrs = FilterGTFChromosomes()
filt_ensembl_gtf_chrs.input = gunzip_ensembl_gtf.output
filt_ensembl_gtf_chrs.output = stripsuffix(gunzip_ensembl_gtf.output,
".gtf") + ".filtered.gtf"
self.add(filt_ensembl_gtf_chrs)
gtf2genepred_ensembl = GTF2GenePred()
gtf2genepred_ensembl.input = filt_ensembl_gtf_chrs.output
gtf2genepred_ensembl.output = stripsuffix(filt_ensembl_gtf_chrs.output,
".gtf") + ".genepred"
self.add(gtf2genepred_ensembl)
filt_genes_ensembl_gtf_genes = FilterGTFGenes()
filt_genes_ensembl_gtf_genes.input = filt_ensembl_gtf_chrs.output
filt_genes_ensembl_gtf_genes.output = stripsuffix(
filt_ensembl_gtf_chrs.output, ".gtf") + ".genes-only.gtf"
self.add(filt_genes_ensembl_gtf_genes)
self.reference_data['ensemblVersion'] = self.ensembl_version
self.reference_data['genesGtf'] = filt_ensembl_gtf_chrs.output
self.reference_data['genesGenePred'] = gtf2genepred_ensembl.output
self.reference_data[
'genesGtfGenesOnly'] = filt_genes_ensembl_gtf_genes.output
def prepare_reference_genome(self):
genome_unzipped = stripsuffix(
os.path.basename(self.input_reference_sequence), ".gz")
gunzip_ref = Gunzip()
gunzip_ref.input = self.input_reference_sequence
gunzip_ref.output = "{}/genome/{}".format(self.outdir, genome_unzipped)
self.add(gunzip_ref)
copy_ref_to_bwa = Copy(input_file=gunzip_ref.output,
output_file="{}/bwa/{}".format(
self.outdir,
os.path.basename(gunzip_ref.output)))
self.add(copy_ref_to_bwa)
bwa_index = BwaIndex()
bwa_index.input_fasta = copy_ref_to_bwa.output
bwa_index.output = copy_ref_to_bwa.output + ".bwt"
bwa_index.algorithm = "bwtsw"
self.add(bwa_index)
create_dict = PicardCreateSequenceDictionary()
create_dict.input = gunzip_ref.output
create_dict.output_dict = gunzip_ref.output.replace(".fasta",
"") + ".dict"
self.add(create_dict)
samtools_faidx = SamtoolsFaidx()
samtools_faidx.input_fasta = gunzip_ref.output
samtools_faidx.output = gunzip_ref.output + ".fai"
self.add(samtools_faidx)
create_chrsizes = GenerateChrSizes()
create_chrsizes.input_fai = samtools_faidx.output
create_chrsizes.output = gunzip_ref.output.replace(
".fasta", "") + ".chrsizes.txt"
self.add(create_chrsizes)
copy_qdnaseq_bg = Copy(input_file=self.qdnaseq_background,
output_file="{}/genome/{}".format(
self.outdir,
os.path.basename(self.qdnaseq_background)))
self.add(copy_qdnaseq_bg)
self.reference_data['reference_genome'] = gunzip_ref.output
self.reference_data['reference_dict'] = create_dict.output_dict
self.reference_data['chrsizes'] = create_chrsizes.output
self.reference_data['bwaIndex'] = bwa_index.input_fasta
self.reference_data['qdnaseq_background'] = copy_qdnaseq_bg.output
def prepare_intervals(self):
self.reference_data['targets'] = {}
target_intervals_dir = "{}/target_intervals/".format(
self.genome_resources)
input_files = [
f for f in os.listdir(target_intervals_dir)
if f.endswith(".interval_list")
]
scan_for_microsatellites = MsiSensorScan()
scan_for_microsatellites.input_fasta = self.reference_data[
'reference_genome']
scan_for_microsatellites.homopolymers_only = True
scan_for_microsatellites.output = "{}/intervals/msisensor-microsatellites.tsv".format(
self.outdir)
self.add(scan_for_microsatellites)
for f in input_files:
file_full_path = "{}/target_intervals/{}".format(
self.genome_resources, f)
logging.debug("Parsing intervals file {}".format(file_full_path))
capture_name = stripsuffix(f, ".interval_list")
self.reference_data['targets'][capture_name] = {}
copy_file = Copy(input_file=file_full_path,
output_file="{}/intervals/targets/{}".format(
self.outdir,
os.path.basename(file_full_path)))
self.add(copy_file)
slop_interval_list = SlopIntervalList()
slop_interval_list.input = copy_file.output
slop_interval_list.output = stripsuffix(
copy_file.output,
".interval_list") + ".slopped20.interval_list"
self.add(slop_interval_list)
interval_list_to_bed = IntervalListToBed()
interval_list_to_bed.input = slop_interval_list.output
interval_list_to_bed.output = stripsuffix(
slop_interval_list.output, ".interval_list") + ".bed"
self.add(interval_list_to_bed)
intersect_msi = IntersectMsiSites()
intersect_msi.input_msi_sites = scan_for_microsatellites.output
intersect_msi.target_bed = interval_list_to_bed.output
intersect_msi.output_msi_sites = stripsuffix(
interval_list_to_bed.output, ".bed") + ".msisites.tsv"
self.add(intersect_msi)
self.prepare_msings(stripsuffix(file_full_path, ".interval_list"),
capture_name)
self.reference_data['targets'][capture_name][
'blacklist-bed'] = None
blacklist_bed = stripsuffix(file_full_path,
".interval_list") + ".blacklist.bed"
if os.path.exists(blacklist_bed):
blacklist_copy = Copy(
input_file=blacklist_bed,
output_file="{}/intervals/targets/{}".format(
self.outdir,
os.path.basename(blacklist_bed),
))
self.add(blacklist_copy)
self.reference_data['targets'][capture_name][
'blacklist-bed'] = blacklist_copy.output
purecn_targets_file = stripsuffix(file_full_path,
".interval_list") + ".purecn.txt"
if os.path.exists(purecn_targets_file):
copy_purecn_targets = Copy(
input_file=purecn_targets_file,
output_file="{}/intervals/targets/{}".format(
self.outdir, os.path.basename(purecn_targets_file)))
self.add(copy_purecn_targets)
self.reference_data['targets'][capture_name][
'purecn_targets'] = copy_purecn_targets.output
else:
self.reference_data['targets'][capture_name][
'purecn_targets'] = None
self.reference_data['targets'][capture_name][
'targets-interval_list'] = copy_file.output
self.reference_data['targets'][capture_name][
'targets-interval_list-slopped20'] = slop_interval_list.output
self.reference_data['targets'][capture_name][
'targets-bed-slopped20'] = interval_list_to_bed.output
self.reference_data['targets'][capture_name][
'msisites'] = intersect_msi.output_msi_sites
# Find all .cnn files and copy + register them for use in cnv kit:
for f in [
f for f in os.listdir(target_intervals_dir)
if (f.endswith(".cnn") or "cnvkit-fix" in f)
]:
self.prepare_cnvkit(f)