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_sveffect_regions(self):
for regions_name in ["ar_regions", "ts_regions", "fusion_regions"]:
file_full_path = "{}/{}.bed".format(
self.genome_resources,
regions_name,
)
copy_regions = Copy(input_file=file_full_path,
output_file="{}/intervals/{}".format(
self.outdir,
os.path.basename(file_full_path),
))
self.reference_data[regions_name] = copy_regions.output
self.add(copy_regions)
def prepare_msings(self, filename_base, capture_name):
"""
Setup the copying of the relevant msings parameter files, for the given base
filename.
:param filename_base: String suffix for all potential msings parameter files.
:param capture_name: String capture name. Note: this information is also contained
in the filename_base parameter; should be refactored.
"""
for msings_extn in ["baseline", "bed", "msi_intervals"]:
msings_ref_file = filename_base + ".msings." + msings_extn
if os.path.exists(msings_ref_file):
copy_msings_ref = Copy(
input_file=msings_ref_file,
output_file="{}/intervals/targets/{}".format(
self.outdir, os.path.basename(msings_ref_file)))
self.add(copy_msings_ref)
self.reference_data['targets'][capture_name][
'msings-' + msings_extn] = copy_msings_ref.output
else:
self.reference_data['targets'][capture_name][
'msings-' + msings_extn] = None
def prepare_cnvkit(self, cnv_kit_ref_filename):
"""
:param cnv_kit_ref_filename: String name of a cnvkit reference file (either *.cnn
or *.cnvkitref.txt), to be registered in self.ref_data.
"""
file_full_path = "{}/target_intervals/{}".format(
self.genome_resources, cnv_kit_ref_filename)
# Extract the capture+library+sampletype strings:
capture_library_sampletype = cnv_kit_ref_filename.split(".")[:3]
copy_cnvkit_ref = Copy(input_file=file_full_path,
output_file="{}/intervals/targets/{}".format(
self.outdir,
os.path.basename(file_full_path)))
self.add(copy_cnvkit_ref)
capture_name = capture_library_sampletype[0]
library_kit_name = capture_library_sampletype[1]
sample_type = capture_library_sampletype[2]
ref_type = "cnvkit-fix"
if cnv_kit_ref_filename.endswith(("cnn")):
ref_type = "cnvkit-ref"
# FIXME: Ugly; refactor. This registers the cnvkit reference file copy in the autoseq genome dictionary:
if ref_type not in self.reference_data['targets'][capture_name]:
self.reference_data['targets'][capture_name][ref_type] = {}
if library_kit_name not in self.reference_data['targets'][
capture_name][ref_type]:
self.reference_data['targets'][capture_name][ref_type][
library_kit_name] = {}
self.reference_data['targets'][capture_name][ref_type][library_kit_name][sample_type] = \
copy_cnvkit_ref.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)
def prepare_variants(self):
curl_dbsnp = CurlSplitAndLeftAlign()
curl_dbsnp.input_reference_sequence = self.reference_data[
'reference_genome']
curl_dbsnp.input_reference_sequence_fai = self.reference_data[
'reference_genome'] + ".fai"
curl_dbsnp.remote = self.dbsnp_remote
curl_dbsnp.output = "{}/variants/{}".format(
self.outdir, os.path.basename(self.dbsnp_remote))
curl_dbsnp.is_intermediate = True
self.add(curl_dbsnp)
filter_dbsnp = VcfFilter()
filter_dbsnp.input = curl_dbsnp.output
filter_dbsnp.filter = "\"! ( SAO = 3 | SAO = 2 )\""
filter_dbsnp.output = "{}/variants/dbsnp142-germline-only.vcf.gz".format(
self.outdir)
self.add(filter_dbsnp)
curl_cosmic = CurlSplitAndLeftAlign()
curl_cosmic.input_reference_sequence = self.reference_data[
'reference_genome']
curl_cosmic.input_reference_sequence_fai = self.reference_data[
'reference_genome'] + ".fai"
curl_cosmic.remote = "file://" + self.cosmic_vcf
curl_cosmic.output = "{}/variants/{}".format(
self.outdir, os.path.basename(self.cosmic_vcf))
self.add(curl_cosmic)
curl_clinvar = CurlSplitAndLeftAlign()
curl_clinvar.input_reference_sequence = self.reference_data[
'reference_genome']
curl_clinvar.input_reference_sequence_fai = self.reference_data[
'reference_genome'] + ".fai"
curl_clinvar.remote = self.clinvar_remote
curl_clinvar.output = "{}/variants/{}".format(
self.outdir, os.path.basename(self.clinvar_remote))
self.add(curl_clinvar)
curl_exac = CurlSplitAndLeftAlign()
curl_exac.input_reference_sequence = self.reference_data[
'reference_genome']
curl_exac.input_reference_sequence_fai = self.reference_data[
'reference_genome'] + ".fai"
curl_exac.remote = self.exac_remote
curl_exac.output = "{}/variants/{}".format(
self.outdir, os.path.basename(self.exac_remote))
self.add(curl_exac)
curl_icgc = CurlSplitAndLeftAlign()
curl_icgc.input_reference_sequence = self.reference_data[
'reference_genome']
curl_icgc.input_reference_sequence_fai = self.reference_data[
'reference_genome'] + ".fai"
curl_icgc.remote = self.icgc_somatic_remote
curl_icgc.output = "{}/variants/{}".format(
self.outdir,
"icgc_release_20_simple_somatic_mutation.aggregated.vcf.gz")
self.add(curl_icgc)
curl_swegene = CurlSplitAndLeftAlign()
curl_swegene.input_reference_sequence = self.reference_data[
'reference_genome']
curl_swegene.input_reference_sequence_fai = self.reference_data[
'reference_genome'] + ".fai"
curl_swegene.remote = "file://" + self.swegene_common_vcf
curl_swegene.output = "{}/variants/{}".format(
self.outdir, os.path.basename(self.swegene_common_vcf))
self.add(curl_swegene)
copy_thousand_genome = Copy(input_file=self.thousand_genome_vcf,
output_file="{}/variants/{}".format(
self.outdir,
os.path.basename(
self.thousand_genome_vcf)))
self.add(copy_thousand_genome)
copy_mills_and_1000g = Copy(
input_file=self.mills_and_1000g_gold_standard,
output_file="{}/variants/{}".format(
self.outdir,
os.path.basename(self.mills_and_1000g_gold_standard)))
self.add(copy_mills_and_1000g)
copy_brca_exchange = Copy(input_file=self.brca_exchange,
output_file="{}/variants/{}".format(
self.outdir,
os.path.basename(self.brca_exchange)))
self.add(copy_brca_exchange)
copy_oncokb = Copy(input_file=self.oncokb,
output_file="{}/variants/{}".format(
self.outdir, os.path.basename(self.oncokb)))
self.add(copy_oncokb)
self.reference_data['dbSNP'] = filter_dbsnp.output
self.reference_data['cosmic'] = curl_cosmic.output
self.reference_data['exac'] = curl_exac.output
self.reference_data['clinvar'] = curl_clinvar.output
self.reference_data['icgc'] = curl_icgc.output
self.reference_data['swegene_common'] = curl_swegene.output
self.reference_data['1KG'] = copy_thousand_genome.output
self.reference_data[
'Mills_and_1KG_gold_standard'] = copy_mills_and_1000g.output
self.reference_data['brca_exchange'] = copy_brca_exchange.output
self.reference_data['oncokb'] = copy_oncokb.output