Data Processing Pipeline: local indel realignment, mark duplicates and BQSR. Done.

Pacbio pipeline: now all pacbio bams have baq annotated in so running UG is uber fast.

Methods pipeline: minor cosmetic changes.



git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@5253 348d0f76-0448-11de-a6fe-93d51630548a
This commit is contained in:
carneiro 2011-02-16 17:22:30 +00:00
parent b304ced801
commit 50c870cfce
3 changed files with 29 additions and 18 deletions

View File

@ -1,4 +1,4 @@
import org.broadinstitute.sting.queue.extensions.gatk.CommandLineGATK
import org.broadinstitute.sting.gatk.CommandLineGATK
import org.broadinstitute.sting.queue.extensions.gatk._
import org.broadinstitute.sting.queue.QScript
import org.broadinstitute.sting.gatk.phonehome.GATKRunReport

View File

@ -1,7 +1,7 @@
import org.broadinstitute.sting.queue.extensions.gatk._
import org.broadinstitute.sting.queue.extensions.picard.PicardBamJarFunction
import org.broadinstitute.sting.queue.QScript
import org.broadinstitute.sting.queue.extensions.samtools.SamtoolsIndexFunction
class dataProcessing extends QScript {
@ -31,12 +31,16 @@ class dataProcessing extends QScript {
def script = {
// Files generated by the pipeline
val baseName: String = swapExt(qscript.inputBam, ".bam", "").toString()
def cleanedBam: String = baseName + ".cleaned.bam"
def dedupedBam: String = baseName + ".cleaned.dedup.bam"
def metricsFile: String = swapExt(qscript.inputBam, "bam", "metrics").toString()
def targetIntervals: String = baseName + ".indel.intervals"
val cleanedBam: String = baseName + ".cleaned.bam"
val dedupedBam: String = baseName + ".cleaned.dedup.bam"
val metricsFile: String = swapExt(qscript.inputBam, "bam", "metrics").toString()
val targetIntervals: String = baseName + ".indel.intervals"
val recalFile: String = baseName + ".recal.csv"
val recalBam: String = baseName + ".cleaned.dedup.recal.bam"
// Reference sequence, dbsnps and RODs used by the pipeline
val reference: File = new File("/humgen/1kg/reference/human_g1k_v37.fasta")
val dbSNP: File = new File("/humgen/gsa-hpprojects/GATK/data/dbsnp_132_b37.leftAligned.vcf")
val dindelPilotCalls: String = "/humgen/gsa-hpprojects/GATK/data/Comparisons/Unvalidated/1kg.pilot_release.merged.indels.sites.hg19.vcf"
@ -44,16 +48,16 @@ class dataProcessing extends QScript {
val dindelASNCalls: String = "/humgen/1kg/DCC/ftp/technical/working/20110126_dindel_august/ASN.dindel_august_release_merged_pilot1.20110126.sites.vcf.gz"
val dindelEURCalls: String = "/humgen/1kg/DCC/ftp/technical/working/20110126_dindel_august/EUR.dindel_august_release_merged_pilot1.20110126.sites.vcf.gz"
// General arguments to all programs
trait CommandLineGATKArgs extends CommandLineGATK {
this.jarFile = qscript.gatkJar
this.reference_sequence = reference
this.memoryLimit = Some(2)
this.memoryLimit = Some(4)
this.jobTempDir = qscript.outputTmpDir
}
val target = new RealignerTargetCreator with CommandLineGATKArgs
target.memoryLimit = Some(4)
target.input_file :+= qscript.inputBam
target.out = new File(targetIntervals)
target.mismatchFraction = Some(0.0)
@ -69,7 +73,6 @@ class dataProcessing extends QScript {
// 2.) Clean without SW
val clean = new IndelRealigner with CommandLineGATKArgs
clean.memoryLimit = Some(4)
clean.input_file :+= qscript.inputBam
clean.targetIntervals = new File(targetIntervals)
clean.out = new File(cleanedBam)
@ -88,9 +91,9 @@ class dataProcessing extends QScript {
// 3.) Mark Duplicates
val dedup = new PicardBamJarFunction{
@Input(doc="cleaned bam") var clean: File = new File(cleanedBam)
@Output(doc="deduped bam") var dedup: File = new File(dedupedBam)
@Output(doc="deduped bam") var deduped: File = new File(dedupedBam)
override def inputBams = List(clean)
override def outputBam = dedup
override def outputBam = deduped
override def commandLine = super.commandLine + " M=" + metricsFile
sortOrder = null
}
@ -98,9 +101,21 @@ class dataProcessing extends QScript {
dedup.jarFile = qscript.dedupJar
dedup.jobName = baseName + ".dedup"
// val cov = new CountCovariates with CommandLineGATKArgs
val index = new SamtoolsIndexFunction
index.bamFile = new File(dedupedBam)
index.analysisName = baseName + ".index"
val cov = new CountCovariates with CommandLineGATKArgs
cov.rodBind :+= RodBind("dbsnp", "VCF", dbSNP)
cov.covariate ++= List("ReadGroupCovariate", "QualityScoreCovariate", "CycleCovariate", "DinucCovariate")
cov.input_file :+= new File(dedupedBam)
cov.recal_file = new File(recalFile)
add(target, clean, dedup)
val recal = new TableRecalibration with CommandLineGATKArgs
recal.input_file :+= new File (dedupedBam)
recal.recal_file = new File(recalFile)
recal.out = new File(recalBam)
add(target, clean, dedup, index, cov, recal)
}
}

View File

@ -15,10 +15,6 @@ class pbCalling extends QScript {
@Argument(shortName="dataset", doc="selects the datasets to run. If not provided, all datasets will be used", required=false)
var datasets: List[String] = Nil
@Argument(shortName="noBAQ", doc="turns off BAQ calculation", required=false)
var noBAQ: Boolean = false
class Target(
val baseName: String,
@ -169,7 +165,7 @@ class pbCalling extends QScript {
this.stand_emit_conf = Some( if ( t.isLowpass ) { 4.0 } else { 30.0 } )
this.input_file :+= t.bamList
this.out = t.rawVCF
this.baq = Some( if (noBAQ) {org.broadinstitute.sting.utils.baq.BAQ.CalculationMode.OFF} else {org.broadinstitute.sting.utils.baq.BAQ.CalculationMode.RECALCULATE})
this.baq = Some(org.broadinstitute.sting.utils.baq.BAQ.CalculationMode.CALCULATE_AS_NECESSARY)
this.analysisName = t.name + "_UG"
if (t.dbsnpFile.endsWith(".rod"))
this.DBSNP = new File(t.dbsnpFile)