def GATKpreprocessing(infile, outfile):
'''Reorders BAM according to reference fasta and add read groups using
SAMtools, realigns around indels and recalibrates base quality scores
using GATK'''
to_cluster = USECLUSTER
track = P.snip(os.path.basename(infile), ".bam")
tmpdir_gatk = P.getTempDir()
job_memory = PARAMS["gatk_memory"]
genome = "%s/%s.fa" % (PARAMS["bwa_index_dir"], PARAMS["genome"])
outfile1 = outfile.replace(".bqsr", ".readgroups.bqsr")
outfile2 = outfile.replace(".bqsr", ".realign.bqsr")
PipelineExome.GATKReadGroups(infile, outfile1, genome,
PARAMS["readgroup_library"],
PARAMS["readgroup_platform"],
PARAMS["readgroup_platform_unit"])
PipelineExome.GATKIndelRealign(outfile1, outfile2, genome,
PARAMS["gatk_threads"])
IOTools.zapFile(outfile1)
PipelineExome.GATKBaseRecal(outfile2, outfile, genome,
PARAMS["gatk_dbsnp"],
PARAMS["gatk_solid_options"])
IOTools.zapFile(outfile2)
def runMutectReverse(infiles, outfile):
'''Use control as tumor and vis versa to estimate false positive rate'''
infile, normal_panel = infiles
infile_tumour = infile.replace(
PARAMS["sample_control"], PARAMS["sample_tumour"])
basename = P.snip(outfile, "_normal_mutect.vcf")
call_stats_out = basename + "_call_stats.out"
mutect_log = basename + ".log"
basename = P.snip(outfile, ".mutect.reverse.snp.vcf")
call_stats_out = basename + "_call_stats.reverse.out"
coverage_wig_out = basename + "_coverage.reverse.wig"
mutect_log = basename + ".reverse.log"
(cosmic, dbsnp, quality, max_alt_qual, max_alt,
max_fraction, tumor_LOD) = (
PARAMS["mutect_cosmic"], PARAMS["gatk_dbsnp"],
PARAMS["mutect_quality"], PARAMS["mutect_max_alt_qual"],
PARAMS["mutect_max_alt"], PARAMS["mutect_max_fraction"],
PARAMS["mutect_LOD"])
genome = "%s/%s.fa" % (PARAMS["bwa_index_dir"],
PARAMS["genome"])
PipelineExome.mutectSNPCaller(infile, outfile, mutect_log, genome,
cosmic, dbsnp, call_stats_out,
PARAMS['mutect_memory'],
PARAMS['mutect_threads'],
quality, max_alt_qual,
max_alt, max_fraction, tumor_LOD,
normal_panel, infile_tumour)
def defineEBioStudies(outfile):
''' For the cancer types specified in pipeline.ini, identify the
relevent studies in eBio '''
cancer_types = PARAMS["annotation_ebio_cancer_types"]
PipelineExome.defineEBioStudies(cancer_types, outfile, submit=False)
def runMutectOnDownsampled(infiles, outfile):
'''call somatic SNPs using MuTect on downsampled bams'''
infile, normal_panel = infiles
infile_tumour = infile.replace(
PARAMS["sample_control"], PARAMS["sample_tumour"])
basename = P.snip(outfile, "_normal_mutect.vcf")
call_stats_out = basename + "_call_stats.out"
mutect_log = basename + ".log"
(cosmic, dbsnp, quality, max_alt_qual, max_alt,
max_fraction, tumor_LOD) = (
PARAMS["mutect_cosmic"], PARAMS["gatk_dbsnp"],
PARAMS["mutect_quality"], PARAMS["mutect_max_alt_qual"],
PARAMS["mutect_max_alt"], PARAMS["mutect_max_fraction"],
PARAMS["mutect_LOD"])
genome = "%s/%s.fa" % (PARAMS["bwa_index_dir"],
PARAMS["genome"])
PipelineExome.mutectSNPCaller(infile_tumour, outfile, mutect_log, genome,
cosmic, dbsnp, call_stats_out,
PARAMS['mutect_memory'], PARAMS['mutect_threads'],
quality, max_alt_qual,
max_alt, max_fraction, tumor_LOD,
normal_panel, infile)
def defineEBioStudies(outfile):
''' For the cancer types specified in pipeline.ini, identify the
relevent studies in eBio '''
cancer_types = PARAMS["annotation_ebio_cancer_types"]
PipelineExome.defineEBioStudies(cancer_types, outfile, submit=False)
def runMutectReverse(infiles, outfile):
'''Use control as tumor and vis versa to estimate false positive rate'''
infile, normal_panel = infiles
infile_tumour = infile.replace(PARAMS["sample_control"],
PARAMS["sample_tumour"])
basename = P.snip(outfile, "_normal_mutect.vcf")
call_stats_out = basename + "_call_stats.out"
mutect_log = basename + ".log"
basename = P.snip(outfile, ".mutect.reverse.snp.vcf")
call_stats_out = basename + "_call_stats.reverse.out"
coverage_wig_out = basename + "_coverage.reverse.wig"
mutect_log = basename + ".reverse.log"
(cosmic, dbsnp, quality, max_alt_qual, max_alt, max_fraction,
tumor_LOD) = (PARAMS["mutect_cosmic"], PARAMS["gatk_dbsnp"],
PARAMS["mutect_quality"], PARAMS["mutect_max_alt_qual"],
PARAMS["mutect_max_alt"], PARAMS["mutect_max_fraction"],
PARAMS["mutect_LOD"])
genome = "%s/%s.fa" % (PARAMS["bwa_index_dir"], PARAMS["genome"])
PipelineExome.mutectSNPCaller(infile, outfile, mutect_log, genome, cosmic,
dbsnp, call_stats_out,
PARAMS['mutect_memory'],
PARAMS['mutect_threads'], quality,
max_alt_qual, max_alt, max_fraction,
tumor_LOD, normal_panel, infile_tumour)
def extractEBioinfo(infiles, outfile):
'''find the number of mutations identitified in previous studies (ebio_ids)
for the mutated genes in the annotated vcfs'''
eBio_ids = infiles[0]
vcfs = infiles[1:]
PipelineExome.extractEBioinfo(eBio_ids, vcfs, outfile, submit=False)
def extractEBioinfo(infiles, outfile):
'''find the number of mutations identitified in previous studies (ebio_ids)
for the mutated genes in the annotated vcfs'''
eBio_ids = infiles[0]
vcfs = infiles[1:]
PipelineExome.extractEBioinfo(eBio_ids, vcfs, outfile, submit=False)
def mutationalSignature(infiles, outfiles):
min_t_alt = PARAMS["filter_minimum_tumor_allele"]
min_t_alt_freq = PARAMS["filter_minimum_tumor_allele_frequency"]
min_n_depth = PARAMS["filter_minimum_normal_depth"]
max_n_alt_freq = PARAMS["filter_maximum_normal_allele_frequency"]
tumour = PARAMS["mutect_tumour"]
PipelineExome.compileMutationalSignature(
infiles, outfiles, min_t_alt, min_n_depth, max_n_alt_freq,
min_t_alt_freq, tumour, submit=True)
def indelCaller(infile, outfile):
'''Call somatic indels using Strelka'''
infile_tumour = infile.replace("Control", PARAMS["mutect_tumour"])
outdir = "/".join(outfile.split("/")[0:2])
genome = "%s/%s.fa" % (PARAMS["bwa_index_dir"],
PARAMS["genome"])
config = "config.ini"
PipelineExome.strelkaINDELCaller(infile, infile_tumour, outfile,
genome, config, outdir,
PARAMS['strelka_memory'],
PARAMS['strelka_threads'])
def indelCaller(infile, outfile):
'''Call somatic indels using Strelka'''
infile_tumour = infile.replace(PARAMS["sample_control"],
PARAMS["sample_tumour"])
outdir = "/".join(outfile.split("/")[0:2])
genome = "%s/%s.fa" % (PARAMS["bwa_index_dir"], PARAMS["genome"])
PipelineExome.strelkaINDELCaller(infile, infile_tumour, outfile, genome,
PARAMS['strelka_config'], outdir,
PARAMS['strelka_memory'],
PARAMS['strelka_threads'])
def filterMutect(infile, outfile):
''' filter mutect snps using allele frequencies '''
logfile = outfile.replace(".vcf", ".log")
min_t_alt = PARAMS["filter_minimum_tumor_allele"]
min_t_alt_freq = PARAMS["filter_minimum_tumor_allele_frequency"]
min_n_depth = PARAMS["filter_minimum_normal_depth"]
max_n_alt_freq = PARAMS["filter_maximum_normal_allele_frequency"]
min_ratio = PARAMS["filter_minimum_ratio"]
PipelineExome.filterMutect(infile, outfile, logfile,
PARAMS["sample_control"],
PARAMS["sample_tumour"], min_t_alt, min_n_depth,
max_n_alt_freq, min_t_alt_freq, min_ratio)
def filterMutect(infile, outfile):
''' filter mutect snps using allele frequencies '''
logfile = outfile.replace(".vcf", ".log")
min_t_alt = PARAMS["filter_minimum_tumor_allele"]
min_t_alt_freq = PARAMS["filter_minimum_tumor_allele_frequency"]
min_n_depth = PARAMS["filter_minimum_normal_depth"]
max_n_alt_freq = PARAMS["filter_maximum_normal_allele_frequency"]
min_ratio = PARAMS["filter_minimum_ratio"]
PipelineExome.filterMutect(
infile, outfile, logfile,
PARAMS["sample_control"], PARAMS["sample_tumour"],
min_t_alt, min_n_depth, max_n_alt_freq,
min_t_alt_freq, min_ratio)
def callControlVariants(infile, outfile):
'''run mutect to call snps in control sample'''
basename = P.snip(outfile, "_normal_mutect.vcf")
call_stats_out = basename + "_call_stats.out"
mutect_log = basename + ".log"
cosmic, dbsnp, = (PARAMS["mutect_cosmic"],
PARAMS["gatk_dbsnp"])
genome = "%s/%s.fa" % (PARAMS["bwa_index_dir"],
PARAMS["genome"])
PipelineExome.mutectSNPCaller(infile, outfile, mutect_log, genome, cosmic,
dbsnp, call_stats_out, PARAMS['mutect_memory'],
PARAMS['mutect_threads'], artifact=True)
def realignMatchedSample(infile, outfile):
''' repeat realignments with merged bam of control and tumor
this should help avoid problems with sample-specific realignments'''
genome = "%s/%s.fa" % (PARAMS["bwa_index_dir"], PARAMS["genome"])
PipelineExome.GATKIndelRealign(infile, outfile, genome)
IOTools.zapFile(infile)
def callControlVariants(infile, outfile):
'''run mutect to call snps in control sample'''
basename = P.snip(outfile, "_normal_mutect.vcf")
call_stats_out = basename + "_call_stats.out"
mutect_log = basename + ".log"
cosmic, dbsnp, = (PARAMS["mutect_cosmic"], PARAMS["gatk_dbsnp"])
genome = "%s/%s.fa" % (PARAMS["bwa_index_dir"], PARAMS["genome"])
PipelineExome.mutectSNPCaller(infile,
outfile,
mutect_log,
genome,
cosmic,
dbsnp,
call_stats_out,
PARAMS['mutect_memory'],
PARAMS['mutect_threads'],
artifact=True)
def runMutectOnDownsampled(infiles, outfile):
'''call somatic SNPs using MuTect on downsampled bams'''
infile, normal_panel = infiles
infile_tumour = infile.replace(PARAMS["sample_control"],
PARAMS["sample_tumour"])
basename = P.snip(outfile, "_normal_mutect.vcf")
call_stats_out = basename + "_call_stats.out"
mutect_log = basename + ".log"
(cosmic, dbsnp, quality, max_alt_qual, max_alt, max_fraction,
tumor_LOD) = (PARAMS["mutect_cosmic"], PARAMS["gatk_dbsnp"],
PARAMS["mutect_quality"], PARAMS["mutect_max_alt_qual"],
PARAMS["mutect_max_alt"], PARAMS["mutect_max_fraction"],
PARAMS["mutect_LOD"])
genome = "%s/%s.fa" % (PARAMS["bwa_index_dir"], PARAMS["genome"])
PipelineExome.mutectSNPCaller(infile_tumour, outfile, mutect_log, genome,
cosmic, dbsnp, call_stats_out,
PARAMS['mutect_memory'],
PARAMS['mutect_threads'], quality,
max_alt_qual, max_alt, max_fraction,
tumor_LOD, normal_panel, infile)
def intersectHeatmap(infiles, outfile):
''' intersect DE test_ids across the different quantifiers'''
PipelineExome.intersectionHeatmap(infiles, outfile)
def mutationalSignature(infiles, outfiles):
PipelineExome.compileMutationalSignature(
infiles, outfiles)
def summariseFiltering(infile, outfile):
infile = infile.replace(".mutect.snp.vcf", "_call_stats.out")
PipelineExome.parseMutectCallStats(infile, outfile, submit=True)
def mutationalSignature(infiles, outfiles):
PipelineExome.compileMutationalSignature(infiles, outfiles)
def intersectHeatmap(infiles, outfile):
''' intersect DE test_ids across the different quantifiers'''
PipelineExome.intersectionHeatmap(infiles, outfile)
def summariseFiltering(infile, outfile):
infile = infile.replace(".mutect.snp.vcf", "_call_stats.out")
PipelineExome.parseMutectCallStats(infile, outfile, submit=True)