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

updated what constitutes removing entropy 

git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@1113 348d0f76-0448-11de-a6fe-93d51630548a
This commit is contained in:
ebanks 2009-06-26 18:29:00 +00:00
parent d7d4298917
commit eb74b16e39
1 changed files with 14 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
java/src/org/broadinstitute/sting/playground/gatk/walkers/indels/IntervalCleanerWalker.java
View File

@ -40,6 +40,9 @@ public class IntervalCleanerWalker extends LocusWindowWalker<Integer, Integer>
public static final int MAX_QUAL = 99; public static final int MAX_QUAL = 99;
// fraction of mismatches that need to no longer mismatch for a column to be considered cleaned
private static final double MISMATCH_COLUMN_CLEANED_FRACTION = 0.75;
private SAMFileWriter writer = null; private SAMFileWriter writer = null;
private FileWriter indelOutput = null; private FileWriter indelOutput = null;
private FileWriter statsOutput = null; private FileWriter statsOutput = null;
@ -577,22 +580,25 @@ public class IntervalCleanerWalker extends LocusWindowWalker<Integer, Integer>
for ( int i=0; i < reference.length(); i++ ) { for ( int i=0; i < reference.length(); i++ ) {
if ( cleanedMismatchBases[i] == originalMismatchBases[i] ) if ( cleanedMismatchBases[i] == originalMismatchBases[i] )
continue; continue;
if ( originalMismatchBases[i] > totalBases[i] * MISMATCH_THRESHOLD ) boolean didMismatch = false, stillMismatches = false;
originalMismatchColumns++;
if ( cleanedMismatchBases[i] > totalBases[i] * MISMATCH_THRESHOLD )
cleanedMismatchColumns++;
if ( snpsOutput != null ) {
if ( originalMismatchBases[i] > totalBases[i] * MISMATCH_THRESHOLD ) { if ( originalMismatchBases[i] > totalBases[i] * MISMATCH_THRESHOLD ) {
didMismatch = true;
originalMismatchColumns++;
if ( cleanedMismatchBases[i] > originalMismatchBases[i] * (1.0 - MISMATCH_COLUMN_CLEANED_FRACTION) ) {
stillMismatches = true;
cleanedMismatchColumns++;
}
} else if ( cleanedMismatchBases[i] > totalBases[i] * MISMATCH_THRESHOLD ) {
cleanedMismatchColumns++;
}
if ( snpsOutput != null ) {
if ( didMismatch ) {
sb.append(reads.get(0).getRead().getReferenceName() + ":"); sb.append(reads.get(0).getRead().getReferenceName() + ":");
sb.append(((int)leftmostIndex + i)); sb.append(((int)leftmostIndex + i));
if ( cleanedMismatchBases[i] > totalBases[i] * MISMATCH_THRESHOLD ) if ( stillMismatches )
sb.append(" SAME_SNP\n"); sb.append(" SAME_SNP\n");
else else
sb.append(" NOT_SNP\n"); sb.append(" NOT_SNP\n");
//} else if ( cleanedMismatchBases[i] > totalBases[i] * MISMATCH_THRESHOLD ) {
// sb.append(reads.get(0).getRead().getReferenceName() + ":");
// sb.append(((int)leftmostIndex + i));
// sb.append(" NEW_SNP\n");
} }
} }
} }