def build_sampling_nodes(options, genotyping, sample, regions, dependencies): fasta_file = regions["Genotypes"][sample] pileup_file = swap_ext(fasta_file, ".pileup.bgz") padding = genotyping["Padding"] slop, node = build_regions_nodes(regions, padding, dependencies) bam_file = "%s.%s.bam" % (sample, regions["Prefix"]) bam_file = os.path.join(options.samples_root, bam_file) if regions["Realigned"]: bam_file = add_postfix(bam_file, ".realigned") bai_node = build_bam_index_node(bam_file) genotype = GenotypeRegionsNode.customize(pileup_only=True, reference=regions["FASTA"], bedfile=slop, infile=bam_file, outfile=pileup_file, nbatches=options.samtools_max_threads, dependencies=node + (bai_node,)) apply_samtools_options(genotype.command, genotyping["MPileup"], "--mpileup-argument") genotype = genotype.build_node() tabix = TabixIndexNode(infile=pileup_file, preset="pileup", dependencies=genotype) builder = SampleRegionsNode(infile=pileup_file, bedfile=regions["BED"], outfile=fasta_file, dependencies=tabix) faidx = FastaIndexNode(infile=fasta_file, dependencies=builder) return (faidx,)
def build_sampling_nodes(options, genotyping, sample, regions, dependencies): fasta_file = regions["Genotypes"][sample] pileup_file = swap_ext(fasta_file, ".pileup.bgz") padding = genotyping["Padding"] slop, node = build_regions_nodes(regions, padding, dependencies) bam_file = "%s.%s.bam" % (sample, regions["Prefix"]) bam_file = os.path.join(options.samples_root, bam_file) if regions["Realigned"]: bam_file = add_postfix(bam_file, ".realigned") bai_node = build_bam_index_node(bam_file) genotype = GenotypeRegionsNode.customize( pileup_only=True, reference=regions["FASTA"], bedfile=slop, infile=bam_file, outfile=pileup_file, nbatches=options.samtools_max_threads, dependencies=node + (bai_node, )) apply_samtools_options(genotype.command, genotyping["MPileup"], "--mpileup-argument") genotype = genotype.build_node() tabix = TabixIndexNode(infile=pileup_file, preset="pileup", dependencies=genotype) builder = SampleRegionsNode(infile=pileup_file, bedfile=regions["BED"], outfile=fasta_file, dependencies=tabix) faidx = FastaIndexNode(infile=fasta_file, dependencies=builder) return (faidx, )
def build_genotyping_nodes_cached(options, genotyping, sample, regions, dependencies): """Carries out genotyping, filtering of calls, and indexing of files for a given sample and prefix. If the option 'GenotypeEntirePrefix' is enabled, the BAM is genotyped once, and each set of RegionsOfInterest simply extract the relevant regions during construction of the consensus sequence. Parameters: options: An options object (c.f. pypeline.tools.phylo_pipeline.config). genotyping: Genotyping options defined for a specific set of areas of interest, corresponding to Genotyping:NAME in the makefile. sample: The name of the sample to be genotyped. egions: A dictionary for a 'RegionsOfInterest' from the makefile. dependencies: Depenencies that must be met before genotyping starts. Returns a tuple containing the filename of the filtered and tabix-indexed VCF file, and the top-level node generating this file. Multiple calls for the same BAM and prefix will return the same VCF and nodes if the option for 'GenotypeEntirePrefix' is enabled, otherwise each ROI is genotyped individiually. Output files are generated in ./results/PROJECT/genotyping. If the option for 'GenotypeEntirePrefix' is enabled, the following files are generated: SAMPLE.PREFIX.vcf.bgz: Unfiltered calls for variant/non-variant sites. SAMPLE.PREFIX.vcf.pileup.bgz: Pileup of sites containing SNPs. SAMPLE.PREFIX.vcf.pileup.bgz.tbi: Tabix index of the pileup. SAMPLE.PREFIX.filtered.vcf.bgz: Variant calls filtered with vcf_filter. SAMPLE.PREFIX.filtered.vcf.bgz.tbi: Tabix index for the filtered VCF. If 'GenotypeEntirePrefix' is not enabled for a given ROI, the following files are generated for that ROI (see descriptions above): SAMPLE.PREFIX.ROI.filtered.vcf.bgz SAMPLE.PREFIX.ROI.filtered.vcf.bgz.tbi SAMPLE.PREFIX.ROI.vcf.bgz SAMPLE.PREFIX.ROI.vcf.pileup.bgz SAMPLE.PREFIX.ROI.vcf.pileup.bgz.tbi In addition, the following files are generated for each set of RegionsOfInterest (ROI), regardless of the 'GenotypeEntirePrefix' option: SAMPLE.PREFIX.ROI.CDS.fasta: FASTA sequence of each feature in the ROI. SAMPLE.PREFIX.ROI.CDS.fasta.fai: FASTA index generated using SAMTools. """ output_prefix, bamfile, bedfile, dependencies \ = build_genotyping_bedfile_nodes(options, genotyping, sample, regions, dependencies) if (bamfile, output_prefix) in _VCF_CACHE: return _VCF_CACHE[(bamfile, output_prefix)] calls = swap_ext(output_prefix, ".vcf.bgz") pileups = swap_ext(output_prefix, ".vcf.pileup.bgz") filtered = swap_ext(output_prefix, ".filtered.vcf.bgz") # 1. Call samtools mpilup | bcftools view on the bam genotype = GenotypeRegionsNode.customize(reference=regions["FASTA"], bedfile=bedfile, infile=bamfile, outfile=calls, nbatches=options.samtools_max_threads, dependencies=dependencies) genotype.command.add_option("--mpileup-argument", "-f=%s" % (regions["FASTA"],), sep="=") apply_samtools_options(genotype.command, genotyping["MPileup"], "--mpileup-argument") apply_samtools_options(genotype.command, genotyping["BCFTools"], "--bcftools-argument") genotype = genotype.build_node() # 2. Collect pileups of sites with SNPs, to allow proper filtering by # frequency of the minor allele, as only the major non-ref allele is # counted in the VCF (c.f. field DP4). vcfpileup = VCFPileupNode.customize(reference=regions["FASTA"], infile_bam=bamfile, infile_vcf=calls, outfile=pileups, dependencies=genotype) apply_samtools_options(vcfpileup.command, genotyping["MPileup"], "--mpileup-argument") vcfpileup = vcfpileup.build_node() vcf_tabix = TabixIndexNode(infile=pileups, preset="pileup", dependencies=vcfpileup) # 3. Filter all sites using the 'vcf_filter' command vcffilter = VCFFilterNode.customize(infile=calls, pileup=pileups, outfile=filtered, regions=regions, dependencies=vcf_tabix) vcffilter = _apply_vcf_filter_options(vcffilter, genotyping, sample) # 4. Tabix index. This allows random-access to the VCF file when building # the consensus FASTA sequence later in the pipeline. tabix = TabixIndexNode(infile=filtered, preset="vcf", dependencies=vcffilter) _VCF_CACHE[(bamfile, output_prefix)] = (filtered, tabix) return filtered, tabix
def build_genotyping_nodes_cached(options, genotyping, sample, regions, dependencies): """Carries out genotyping, filtering of calls, and indexing of files for a given sample and prefix. If the option 'GenotypeEntirePrefix' is enabled, the BAM is genotyped once, and each set of RegionsOfInterest simply extract the relevant regions during construction of the consensus sequence. Parameters: options: An options object (c.f. pypeline.tools.phylo_pipeline.config). genotyping: Genotyping options defined for a specific set of areas of interest, corresponding to Genotyping:NAME in the makefile. sample: The name of the sample to be genotyped. egions: A dictionary for a 'RegionsOfInterest' from the makefile. dependencies: Depenencies that must be met before genotyping starts. Returns a tuple containing the filename of the filtered and tabix-indexed VCF file, and the top-level node generating this file. Multiple calls for the same BAM and prefix will return the same VCF and nodes if the option for 'GenotypeEntirePrefix' is enabled, otherwise each ROI is genotyped individiually. Output files are generated in ./results/PROJECT/genotyping. If the option for 'GenotypeEntirePrefix' is enabled, the following files are generated: SAMPLE.PREFIX.vcf.bgz: Unfiltered calls for variant/non-variant sites. SAMPLE.PREFIX.vcf.pileup.bgz: Pileup of sites containing SNPs. SAMPLE.PREFIX.vcf.pileup.bgz.tbi: Tabix index of the pileup. SAMPLE.PREFIX.filtered.vcf.bgz: Variant calls filtered with vcf_filter. SAMPLE.PREFIX.filtered.vcf.bgz.tbi: Tabix index for the filtered VCF. If 'GenotypeEntirePrefix' is not enabled for a given ROI, the following files are generated for that ROI (see descriptions above): SAMPLE.PREFIX.ROI.filtered.vcf.bgz SAMPLE.PREFIX.ROI.filtered.vcf.bgz.tbi SAMPLE.PREFIX.ROI.vcf.bgz SAMPLE.PREFIX.ROI.vcf.pileup.bgz SAMPLE.PREFIX.ROI.vcf.pileup.bgz.tbi In addition, the following files are generated for each set of RegionsOfInterest (ROI), regardless of the 'GenotypeEntirePrefix' option: SAMPLE.PREFIX.ROI.CDS.fasta: FASTA sequence of each feature in the ROI. SAMPLE.PREFIX.ROI.CDS.fasta.fai: FASTA index generated using SAMTools. """ output_prefix, bamfile, bedfile, dependencies \ = build_genotyping_bedfile_nodes(options, genotyping, sample, regions, dependencies) if (bamfile, output_prefix) in _VCF_CACHE: return _VCF_CACHE[(bamfile, output_prefix)] calls = swap_ext(output_prefix, ".vcf.bgz") pileups = swap_ext(output_prefix, ".vcf.pileup.bgz") filtered = swap_ext(output_prefix, ".filtered.vcf.bgz") # 1. Call samtools mpilup | bcftools view on the bam genotype = GenotypeRegionsNode.customize( reference=regions["FASTA"], bedfile=bedfile, infile=bamfile, outfile=calls, nbatches=options.samtools_max_threads, dependencies=dependencies) apply_samtools_options(genotype.command, genotyping["MPileup"], "--mpileup-argument") apply_samtools_options(genotype.command, genotyping["BCFTools"], "--bcftools-argument") genotype = genotype.build_node() # 2. Collect pileups of sites with SNPs, to allow proper filtering by # frequency of the minor allele, as only the major non-ref allele is # counted in the VCF (c.f. field DP4). vcfpileup = VCFPileupNode.customize(reference=regions["FASTA"], infile_bam=bamfile, infile_vcf=calls, outfile=pileups, dependencies=genotype) apply_samtools_options(vcfpileup.command, genotyping["MPileup"], "--mpileup-argument") vcfpileup = vcfpileup.build_node() vcf_tabix = TabixIndexNode(infile=pileups, preset="pileup", dependencies=vcfpileup) # 3. Filter all sites using the 'vcf_filter' command vcffilter = VCFFilterNode.customize(infile=calls, pileup=pileups, outfile=filtered, regions=regions, dependencies=vcf_tabix) vcffilter = _apply_vcf_filter_options(vcffilter, genotyping, sample) # 4. Tabix index. This allows random-access to the VCF file when building # the consensus FASTA sequence later in the pipeline. tabix = TabixIndexNode(infile=filtered, preset="vcf", dependencies=vcffilter) _VCF_CACHE[(bamfile, output_prefix)] = (filtered, tabix) return filtered, tabix