use of com.hartwig.pipeline.execution.vm.OutputUpload in project pipeline5 by hartwigmedical.
the class RnaArriba method execute.
@Override
public VirtualMachineJobDefinition execute(InputBundle inputs, RuntimeBucket bucket, BashStartupScript startupScript, RuntimeFiles executionFlags) {
InputFileDescriptor descriptor = inputs.get();
final String batchInputs = descriptor.inputValue();
final String sampleId = batchInputs;
// copy down BAM and index file for this sample
final String bamFile = String.format("%s%s", sampleId, RNA_BAM_FILE_ID);
startupScript.addCommand(() -> format("gsutil -u hmf-crunch cp %s/%s/%s %s", RNA_COHORT_LOCATION_V37, sampleId, bamFile, VmDirectories.INPUT));
final String bamIndexFile = String.format("%s%s", sampleId, RNA_BAM_INDEX_FILE_ID);
startupScript.addCommand(() -> format("gsutil -u hmf-crunch cp %s/%s/%s %s", RNA_COHORT_LOCATION_V37, sampleId, bamIndexFile, VmDirectories.INPUT));
// copy down the executable
startupScript.addCommand(() -> format("gsutil -u hmf-crunch cp %s/%s %s", ARRIBA_RESOURCES, ARRIBA_TOOL, VmDirectories.TOOLS));
startupScript.addCommand(() -> format("chmod a+x %s/%s", VmDirectories.TOOLS, ARRIBA_TOOL));
// copy down required reference files
startupScript.addCommand(() -> format("gsutil -u hmf-crunch cp -r %s/%s %s", ARRIBA_RESOURCES, REF_GENOME, VmDirectories.INPUT));
startupScript.addCommand(() -> format("gsutil -u hmf-crunch cp %s/%s %s", ARRIBA_RESOURCES, GENE_DEFINITIONS, VmDirectories.INPUT));
startupScript.addCommand(() -> format("gsutil -u hmf-crunch cp %s/%s %s", ARRIBA_RESOURCES, BLACKLIST, VmDirectories.INPUT));
startupScript.addCommand(() -> format("cd %s", VmDirectories.OUTPUT));
// run Arriba
StringBuilder arribaArgs = new StringBuilder();
arribaArgs.append(String.format(" -x %s/%s", VmDirectories.INPUT, bamFile));
arribaArgs.append(String.format(" -o %s/%s.fusions.tsv", VmDirectories.OUTPUT, sampleId));
arribaArgs.append(String.format(" -O %s/%s.fusions.discarded.tsv", VmDirectories.OUTPUT, sampleId));
arribaArgs.append(String.format(" -a %s/%s", VmDirectories.INPUT, REF_GENOME));
arribaArgs.append(String.format(" -g %s/%s", VmDirectories.INPUT, GENE_DEFINITIONS));
arribaArgs.append(String.format(" -b %s/%s", VmDirectories.INPUT, BLACKLIST));
arribaArgs.append(" -T -P");
startupScript.addCommand(() -> format("%s/%s %s", VmDirectories.TOOLS, ARRIBA_TOOL, arribaArgs.toString()));
/*
./tools/arriba_v1.1.0/arriba
-x ./runs/CPCT02020378T/CPCT02020378T.sorted.bam
-o ./runs/CPCT02020378T/fusions.tsv -O
./runs/CPCT02020378T/fusions.discarded.tsv
-a "./ref/hs37d5_GENCODE19/hs37d5.fa"
-g "./ref/hs37d5_GENCODE19/GENCODE19.gtf"
-b "./tools/arriba_v1.1.0/database/blacklist_hg19_hs37d5_GRCh37_2018-11-04.tsv.gz"
-T -P
*/
// upload the results
startupScript.addCommand(new OutputUpload(GoogleStorageLocation.of(bucket.name(), "arriba"), executionFlags));
// copy results to rna-analysis location on crunch
startupScript.addCommand(() -> format("gsutil -m cp %s/* %s/%s/arriba/", VmDirectories.OUTPUT, RNA_COHORT_LOCATION_V37, sampleId));
return ImmutableVirtualMachineJobDefinition.builder().name("rna-arriba").startupCommand(startupScript).namespacedResults(ResultsDirectory.defaultDirectory()).workingDiskSpaceGb(100).performanceProfile(VirtualMachinePerformanceProfile.custom(12, 64)).build();
}
use of com.hartwig.pipeline.execution.vm.OutputUpload in project pipeline5 by hartwigmedical.
the class RnaIsofoxExonCounts method execute.
@Override
public VirtualMachineJobDefinition execute(InputBundle inputs, RuntimeBucket bucket, BashStartupScript startupScript, RuntimeFiles executionFlags) {
InputFileDescriptor descriptor = inputs.get();
final String batchInputs = descriptor.inputValue();
final String[] batchItems = batchInputs.split(",");
if (batchItems.length < 2) {
System.out.print(String.format("invalid input arguments(%s) - expected SampleId,ReadLength", batchInputs));
return null;
}
final String sampleId = batchItems[COL_SAMPLE_ID];
final String geneIds = batchItems[COL_GENE_IDS];
final RefGenomeVersion refGenomeVersion = V37;
final ResourceFiles resourceFiles = buildResourceFiles(refGenomeVersion);
// final String rnaCohortDirectory = getRnaCohortDirectory(refGenomeVersion);
final String samplesDir = String.format("%s/%s", getRnaCohortDirectory(refGenomeVersion), "samples");
// copy down BAM and index file for this sample
final String bamFile = String.format("%s%s", sampleId, RNA_BAM_FILE_ID);
startupScript.addCommand(() -> format("gsutil -u hmf-crunch cp %s/%s/%s %s", samplesDir, sampleId, bamFile, VmDirectories.INPUT));
final String bamIndexFile = String.format("%s%s", sampleId, RNA_BAM_INDEX_FILE_ID);
startupScript.addCommand(() -> format("gsutil -u hmf-crunch cp %s/%s/%s %s", samplesDir, sampleId, bamIndexFile, VmDirectories.INPUT));
// copy down the Isofox JAR
startupScript.addCommand(() -> format("gsutil -u hmf-crunch cp %s/%s %s", ISOFOX_LOCATION, ISOFOX_JAR, VmDirectories.TOOLS));
startupScript.addCommand(() -> format("cd %s", VmDirectories.OUTPUT));
// run Isofox
StringJoiner isofoxArgs = new StringJoiner(" ");
isofoxArgs.add(String.format("-sample %s", sampleId));
isofoxArgs.add(String.format("-functions %s", FUNC_TRANSCRIPT_COUNTS));
isofoxArgs.add(String.format("-output_dir %s/", VmDirectories.OUTPUT));
isofoxArgs.add(String.format("-bam_file %s/%s", VmDirectories.INPUT, bamFile));
isofoxArgs.add(String.format("-ref_genome %s", resourceFiles.refGenomeFile()));
isofoxArgs.add(String.format("-ensembl_data_dir %s", resourceFiles.ensemblDataCache()));
isofoxArgs.add(String.format("-write_exon_data"));
// isofoxArgs.add(String.format("-write_read_data"));
isofoxArgs.add(String.format("-restricted_gene_ids %s", geneIds));
startupScript.addCommand(() -> format("java -jar %s/%s %s", VmDirectories.TOOLS, ISOFOX_JAR, isofoxArgs.toString()));
// upload the results
startupScript.addCommand(new OutputUpload(GoogleStorageLocation.of(bucket.name(), "isofox"), executionFlags));
return ImmutableVirtualMachineJobDefinition.builder().name("rna-isofox").startupCommand(startupScript).namespacedResults(ResultsDirectory.defaultDirectory()).workingDiskSpaceGb(MAX_EXPECTED_BAM_SIZE_GB).build();
}
use of com.hartwig.pipeline.execution.vm.OutputUpload in project pipeline5 by hartwigmedical.
the class RnaIsofoxSpliceJunctions method execute.
@Override
public VirtualMachineJobDefinition execute(InputBundle inputs, RuntimeBucket bucket, BashStartupScript startupScript, RuntimeFiles executionFlags) {
InputFileDescriptor descriptor = inputs.get();
final String batchInputs = descriptor.inputValue();
final String[] batchItems = batchInputs.split(",");
if (batchItems.length < 2) {
System.out.print(String.format("invalid input arguments(%s) - expected SampleId,ReadLength", batchInputs));
return null;
}
final String sampleId = batchItems[COL_SAMPLE_ID];
final String geneIds = batchItems[COL_GENE_IDS];
final RefGenomeVersion refGenomeVersion = V37;
final ResourceFiles resourceFiles = buildResourceFiles(refGenomeVersion);
final String samplesDir = String.format("%s/%s", getRnaCohortDirectory(refGenomeVersion), "samples");
// copy down BAM and index file for this sample
final String bamFile = String.format("%s%s", sampleId, RNA_BAM_FILE_ID);
startupScript.addCommand(() -> format("gsutil -u hmf-crunch cp %s/%s/%s %s", samplesDir, sampleId, bamFile, VmDirectories.INPUT));
final String bamIndexFile = String.format("%s%s", sampleId, RNA_BAM_INDEX_FILE_ID);
startupScript.addCommand(() -> format("gsutil -u hmf-crunch cp %s/%s/%s %s", samplesDir, sampleId, bamIndexFile, VmDirectories.INPUT));
// copy down the executable
startupScript.addCommand(() -> format("gsutil -u hmf-crunch cp %s/%s %s", ISOFOX_LOCATION, ISOFOX_JAR, VmDirectories.TOOLS));
startupScript.addCommand(() -> format("cd %s", VmDirectories.OUTPUT));
// run Isofox
StringJoiner isofoxArgs = new StringJoiner(" ");
isofoxArgs.add(String.format("-sample %s", sampleId));
isofoxArgs.add(String.format("-functions %s", FUNC_TRANSCRIPT_COUNTS));
isofoxArgs.add(String.format("-output_dir %s/", VmDirectories.OUTPUT));
isofoxArgs.add(String.format("-bam_file %s/%s", VmDirectories.INPUT, bamFile));
isofoxArgs.add(String.format("-ref_genome %s", resourceFiles.refGenomeFile()));
isofoxArgs.add(String.format("-ensembl_data_dir %s", resourceFiles.ensemblDataCache()));
isofoxArgs.add(String.format("-write_exon_data"));
isofoxArgs.add(String.format("-restricted_gene_ids %s", geneIds));
isofoxArgs.add(" -output_id gene_sj");
startupScript.addCommand(() -> format("java -jar %s/%s %s", VmDirectories.TOOLS, ISOFOX_JAR, isofoxArgs.toString()));
// upload the results
startupScript.addCommand(new OutputUpload(GoogleStorageLocation.of(bucket.name(), "isofox"), executionFlags));
return ImmutableVirtualMachineJobDefinition.builder().name("rna-isofox").startupCommand(startupScript).namespacedResults(ResultsDirectory.defaultDirectory()).workingDiskSpaceGb(MAX_EXPECTED_BAM_SIZE_GB).build();
}
use of com.hartwig.pipeline.execution.vm.OutputUpload in project pipeline5 by hartwigmedical.
the class RnaRsem method execute.
@Override
public VirtualMachineJobDefinition execute(InputBundle inputs, RuntimeBucket bucket, BashStartupScript startupScript, RuntimeFiles executionFlags) {
InputFileDescriptor descriptor = inputs.get();
final String batchInputs = descriptor.inputValue();
final String[] batchItems = batchInputs.split(",");
if (batchItems.length != 2) {
System.out.print(String.format("invalid input arguments(%s) - expected SampleId,PathToFastqFiles", batchInputs));
return null;
}
final String sampleId = batchItems[0];
final String fastqFilelist = batchItems[1];
final List<String> sampleFastqFiles = getSampleFastqFileList(sampleId, fastqFilelist);
if (sampleFastqFiles.isEmpty()) {
System.out.print(String.format("sampleId(%s) fastq files not found", sampleId));
return null;
}
// copy down FASTQ files for this sample
for (final String fastqFile : sampleFastqFiles) {
startupScript.addCommand(() -> format("gsutil -u hmf-crunch cp %s %s", fastqFile, VmDirectories.INPUT));
}
// download the executables
startupScript.addCommand(() -> format("gsutil -u hmf-crunch cp -r %s/%s %s", RSEM_RESOURCES, RSEM_TOOL, VmDirectories.TOOLS));
startupScript.addCommand(() -> format("chmod a+x %s/%s/*", VmDirectories.TOOLS, RSEM_TOOL));
// locate the FASTQ files for reads 1 and 2
final String r1Files = format("$(ls %s/*_R1* | tr '\\n' ',')", VmDirectories.INPUT);
final String r2Files = format("$(ls %s/*_R2* | tr '\\n' ',')", VmDirectories.INPUT);
// download reference files
startupScript.addCommand(() -> format("gsutil -u hmf-crunch cp -r %s/%s %s", RNA_RESOURCES, REF_GENCODE_37_DIR, VmDirectories.INPUT));
startupScript.addCommand(() -> format("gsutil -u hmf-crunch cp -r %s/%s %s", RSEM_RESOURCES, RSEM_GENE_INDEX_DIR, VmDirectories.INPUT));
startupScript.addCommand(() -> format("cd %s", VmDirectories.OUTPUT));
// run STAR with transcriptome mapping
final String refGenomeDir = String.format("%s/%s", VmDirectories.INPUT, REF_GENCODE_37_DIR);
final String threadCount = Bash.allCpus();
final String[] starArgs = { "--runThreadN", threadCount, "--genomeDir", refGenomeDir, "--genomeLoad", "NoSharedMemory", "--readFilesIn", r1Files, r2Files, "--readFilesCommand", "zcat", "--outSAMtype", "BAM", "Unsorted", "--outSAMunmapped", "Within", "--outBAMcompression", "0", "--outSAMattributes", "All", "--outFilterMultimapNmax", "10", "--outFilterMismatchNmax", "3", "limitOutSJcollapsed", "3000000", "--chimSegmentMin", "10", "--chimOutType", "WithinBAM", "SoftClip", "--chimJunctionOverhangMin", "10", "--chimSegmentReadGapMax", "3", "--chimScoreMin", "1", "--chimScoreDropMax", "30", "--chimScoreJunctionNonGTAG", "0", "--chimScoreSeparation", "1", "--outFilterScoreMinOverLread", "0.33", "--outFilterMatchNminOverLread", "0.33", "--outFilterMatchNmin", "35", "--alignSplicedMateMapLminOverLmate", "0.33", "--alignSplicedMateMapLmin", "35", "--alignSJstitchMismatchNmax", "5", "-1", "5", "5", "--quantMode", // key line for RSEM;
"TranscriptomeSAM" };
startupScript.addCommand(new VersionedToolCommand("star", "STAR", "2.7.3a", starArgs));
// key output file is 'Aligned.toTranscriptome.out.bam'
// ./tools/RSEM-1.3.3/rsem-calculate-expression --alignments --paired-end
// ./runs/CPCT02020378T/Aligned.toTranscriptome.out.bam
// ./ref/rsem_gene_index/human_gencode
// CPCT02020378T.rsem -p 6 &
final String transcriptomeBam = "Aligned.toTranscriptome.out.bam";
// TMP: copy transcriptome BAM to the bucket
// startupScript.addCommand(() -> format("gsutil -m cp %s/%s gs://rna-cohort/%s/rsem/", VmDirectories.OUTPUT, transcriptomeBam, sampleId));
final String rsemGeneIndex = String.format("%s/%s/%s", VmDirectories.INPUT, RSEM_GENE_INDEX_DIR, RSEM_GENE_INDEX);
final String outputPrefix = String.format("%s.rsem", sampleId);
// run RSEM
StringBuilder rsemArgs = new StringBuilder();
rsemArgs.append(" --alignments");
rsemArgs.append(" --paired-end");
rsemArgs.append(String.format(" %s", transcriptomeBam));
rsemArgs.append(String.format(" %s", rsemGeneIndex));
rsemArgs.append(String.format(" %s", outputPrefix));
rsemArgs.append(String.format(" -p %s", threadCount));
// run RSEM transcript expression calcs
startupScript.addCommand(() -> format("%s/%s/%s %s", VmDirectories.TOOLS, RSEM_TOOL, RSEM_EXPRESSION_CMD, rsemArgs.toString()));
startupScript.addCommand(() -> format("mv %s.rsem.genes.results %s.rsem.gene_data.tsv", sampleId, sampleId));
startupScript.addCommand(() -> format("mv %s.rsem.isoforms.results %s.rsem.trans_data.tsv", sampleId, sampleId));
startupScript.addCommand(new OutputUpload(GoogleStorageLocation.of(bucket.name(), "rsem"), executionFlags));
// copy results to rna-analysis location on crunch
startupScript.addCommand(() -> format("gsutil -m cp %s/*tsv %s/%s/rsem/", VmDirectories.OUTPUT, RNA_COHORT_LOCATION_V37, sampleId));
return ImmutableVirtualMachineJobDefinition.builder().name("rna-rsem").startupCommand(startupScript).namespacedResults(ResultsDirectory.defaultDirectory()).workingDiskSpaceGb(500).performanceProfile(VirtualMachinePerformanceProfile.custom(12, 36)).build();
}
use of com.hartwig.pipeline.execution.vm.OutputUpload in project pipeline5 by hartwigmedical.
the class SageCreatePon method execute.
@Override
public VirtualMachineJobDefinition execute(final InputBundle inputs, final RuntimeBucket runtimeBucket, final BashStartupScript startupScript, final RuntimeFiles executionFlags) {
final String output = String.format("%s/SAGE.pon.vcf.gz", VmDirectories.OUTPUT);
final BashCommand sageCommand = new SageCommand("com.hartwig.hmftools.sage.pon.PonApplication", "100G", "-in", VmDirectories.INPUT, "-out", output, "-threads", Bash.allCpus());
// Download required resources
// startupScript.addCommand(() -> format("gsutil -u hmf-crunch cp %s %s",
// "gs://batch-sage-validation/resources/sage.jar",
// "/opt/tools/sage/" + Versions.SAGE + "/sage.jar"));
// Download germline VCFS (and indexes)
startupScript.addCommand(() -> format("gsutil -u hmf-crunch -m cp %s %s", "gs://batch-sage/*/sage/*.sage.somatic.vcf.gz", VmDirectories.INPUT));
startupScript.addCommand(() -> format("gsutil -u hmf-crunch -m cp %s %s", "gs://batch-sage/*/sage/*.sage.somatic.vcf.gz.tbi", VmDirectories.INPUT));
// Run Pon Generator
startupScript.addCommand(sageCommand);
// Store output
startupScript.addCommand(new OutputUpload(GoogleStorageLocation.of(runtimeBucket.name(), "sage"), executionFlags));
return VirtualMachineJobDefinition.sageSomaticCalling(startupScript, ResultsDirectory.defaultDirectory());
}
Aggregations