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

Add option in ValidationAmplicons to only output SNPs and INDELs, ignoring complex variants (or SVs, etc.)

This commit is contained in:
Guillermo del Angel 2012-03-14 14:26:40 -04:00
parent 8e96969744
commit eca055ccad
1 changed files with 13 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
public/java/src/org/broadinstitute/sting/gatk/walkers/validation/ValidationAmplicons.java
View File

@ -117,6 +117,13 @@ public class ValidationAmplicons extends RodWalker<Integer,Integer> {
@Argument(doc="Only output valid sequences.",fullName="onlyOutputValidAmplicons",required=false) @Argument(doc="Only output valid sequences.",fullName="onlyOutputValidAmplicons",required=false)
boolean onlyOutputValidAmplicons = false; boolean onlyOutputValidAmplicons = false;
/**
* If ignoreComplexEvents is true, the output fasta file will contain only sequences coming from SNPs and Indels.
* Complex substitutions will be ignored.
*/
@Argument(doc="Ignore complex genomic records.",fullName="ignoreComplexEvents",required=false)
boolean ignoreComplexEvents = false;
/** /**
* BWA single-end alignment is used as a primer specificity proxy. Low-complexity regions (that don't align back to themselves as a best hit) are lowercased. * BWA single-end alignment is used as a primer specificity proxy. Low-complexity regions (that don't align back to themselves as a best hit) are lowercased.
* This changes the size of the k-mer used for alignment. * This changes the size of the k-mer used for alignment.
@ -146,6 +153,7 @@ public class ValidationAmplicons extends RodWalker<Integer,Integer> {
StringBuilder rawSequence; StringBuilder rawSequence;
boolean sequenceInvalid; boolean sequenceInvalid;
boolean isSiteSNP; boolean isSiteSNP;
boolean isSiteIndel;
List<String> invReason; List<String> invReason;
int indelCounter; int indelCounter;
@ -244,6 +252,7 @@ public class ValidationAmplicons extends RodWalker<Integer,Integer> {
} else if ( validate != null ) { } else if ( validate != null ) {
// record variant type in case it's needed in output format // record variant type in case it's needed in output format
isSiteSNP = (validate.isSNP()); isSiteSNP = (validate.isSNP());
isSiteIndel = (validate.isIndel());
// doesn't matter if there's a mask here too -- this is what we want to validate // doesn't matter if there's a mask here too -- this is what we want to validate
if ( validate.isFiltered() ) { if ( validate.isFiltered() ) {
logger.warn("You are attempting to validate a filtered site. Why are you attempting to validate a filtered site? You should not be attempting to validate a filtered site."); logger.warn("You are attempting to validate a filtered site. Why are you attempting to validate a filtered site? You should not be attempting to validate a filtered site.");
@ -504,6 +513,9 @@ public class ValidationAmplicons extends RodWalker<Integer,Integer> {
} }
if (ignoreComplexEvents && !isSiteIndel && !isSiteSNP)
return;
if (!onlyOutputValidAmplicons || !sequenceInvalid) { if (!onlyOutputValidAmplicons || !sequenceInvalid) {
String seqIdentity = sequence.toString().replace('n', 'N').replace('i', 'I').replace('d', 'D'); String seqIdentity = sequence.toString().replace('n', 'N').replace('i', 'I').replace('d', 'D');
if (sequenomOutput) { if (sequenomOutput) {
@ -512,7 +524,7 @@ public class ValidationAmplicons extends RodWalker<Integer,Integer> {
out.printf("%s_%s %s%n", allelePos != null ? allelePos.toString() : "multiple", probeName, seqIdentity); out.printf("%s_%s %s%n", allelePos != null ? allelePos.toString() : "multiple", probeName, seqIdentity);
} }
else if (ilmnOutput) { else if (ilmnOutput) {
String type = isSiteSNP?"SNP":"INDEL"; String type = isSiteSNP?"SNP":(isSiteIndel?"INDEL":"OTHER");
seqIdentity = seqIdentity.replace("*",""); // no * in ref allele seqIdentity = seqIdentity.replace("*",""); // no * in ref allele
out.printf("%s,%s,%s,%s,%d,37,1000G,ExomePhase1,Forward,Plus,FALSE%n",probeName,type,seqIdentity,allelePos.getContig(),allelePos.getStart()); out.printf("%s,%s,%s,%s,%d,37,1000G,ExomePhase1,Forward,Plus,FALSE%n",probeName,type,seqIdentity,allelePos.getContig(),allelePos.getStart());
} }