zzh
/
gatk-3.8
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Read error rate is now parallelizable 

git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@3447 348d0f76-0448-11de-a6fe-93d51630548a
This commit is contained in:
chartl 2010-05-27 19:00:09 +00:00
parent 99aae4ac0b
commit ff4a0764df
1 changed files with 30 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

31
java/src/org/broadinstitute/sting/oneoffprojects/walkers/ReadErrorRateWalker.java
View File

@ -29,11 +29,13 @@ import org.broadinstitute.sting.gatk.refdata.ReadMetaDataTracker;
import org.broadinstitute.sting.gatk.walkers.ReadWalker;
import org.broadinstitute.sting.gatk.contexts.ReferenceContext;
import org.broadinstitute.sting.commandline.Argument;
import org.broadinstitute.sting.gatk.walkers.TreeReducible;
import org.broadinstitute.sting.utils.QualityUtils;
import org.broadinstitute.sting.utils.BaseUtils;
import net.sf.samtools.SAMRecord;
import java.util.HashMap;
import java.util.Map;
import java.util.Random;
/**
@ -45,7 +47,7 @@ import java.util.Random;
*
* @author Kiran Garimella
*/
public class ReadErrorRateWalker extends ReadWalker<boolean[], ReadErrorRateCollection> {
public class ReadErrorRateWalker extends ReadWalker<boolean[], ReadErrorRateCollection> implements TreeReducible<ReadErrorRateCollection> {
@Argument(fullName="printVisualHits", shortName="v", doc="print visual hits", required=false) public boolean printVisualHits = false;
@Argument(fullName="useNextBestBase", shortName="nb", doc="use next best base", required=false) public boolean useNextBestBase = false;
@Argument(fullName="useNonNextBestBase",shortName="nnb",doc="use nonnext best base",required=false) public boolean useNonNextBestBase = false;
@ -180,6 +182,17 @@ public class ReadErrorRateWalker extends ReadWalker<boolean[], ReadErrorRateColl
return collection;
}
/**
* For multithreading - take two read error rate collections and put them together
* @param left one collection
* @param right another collection
* @return left updated with the counts from right
*/
public ReadErrorRateCollection treeReduce(ReadErrorRateCollection left, ReadErrorRateCollection right) {
left.merge(right);
return left;
}
/**
* We've processed all the reads. Emit the final, normalized error rate data.
*
@ -222,6 +235,16 @@ class ReadErrorRateCollection {
return builder.toString();
}
public void merge(ReadErrorRateCollection other) {
for ( Map.Entry<Integer,int[]> errorCounts : other.readsByReadLength.entrySet() ) {
if ( this.readsByReadLength.keySet().contains(errorCounts.getKey()) ) {
mergeCounts(readsByReadLength.get(errorCounts.getKey()),errorCounts.getValue());
} else {
readsByReadLength.put(errorCounts.getKey(),errorCounts.getValue());
}
}
}
private static int[] zeroArray( int length ) {
int[] array = new int[length];
for ( int ii = 0; ii < length; ii ++ ) {
@ -231,6 +254,12 @@ class ReadErrorRateCollection {
return array;
}
private static void mergeCounts ( int[] addToMe, int[] dontTouchMe ) {
for ( int index = 0; index < addToMe.length; index ++ ) {
addToMe[index] += dontTouchMe[index];
}
}
public static void updateErrorCounts(int[] sum, boolean[] value) {
for (int cycle = 0; cycle < value.length; cycle++) {