MuTect2 ->Reference name for '1283' not found in sequence dictionary

t1000t1000 Member
edited November 2017 in Ask the GATK team

Hi, I am trying to run
java -jar GenomeAnalysisTK.jar -T MuTect2 -R ucsc.hg19.fasta -I:tumor tumor_rg.bam -I:normal normal_rg.bam -o output.vcf

I am using ucsc.hg19.fasta as a reference and I used:
1. samtools faidx to create the ucsc.hg19.fasta.fai file
2. java -jar picard.jar CreateSequenceDictionary REFERENCE=ucsc.hg19.fasta OUTPUT=ucsc.hg19.fasta.dict

I added a ReadGroup to my two BAM Files:
java -jar picard.jar AddOrReplaceReadGroups I=normal.bam O=normal_rg.bam RGID=4 RGLB=lib1 RGPL=illumina RGPU=unit1 RGSM=20

I also validated the BAM files

Now I get the following error:

##### ERROR MESSAGE: SAM/BAM/CRAM file /analysis/tumnormalor.bam is malformed. Error details: Reference name for '1283' not found in sequence dictionary.

If I exchange the tumor and normal .bam files as command line arguments I get the error
##### ERROR MESSAGE: SAM/BAM/CRAM file /analysis/tumor.bam is malformed. Error details: Reference name for '1283' not found in sequence dictionary.

I tried to google it and found nothing. Can someone give me a clue where to look ?

Is the file ucsc.hg19.fasta.dict the sequence dictionary ?

Tagged:

Answers

Sign In or Register to comment.