To celebrate the release of GATK 4.0, we are giving away free credits for running the GATK4 Best Practices pipelines in FireCloud, our secure online analysis portal. It’s first come first serve, so sign up now to claim your free credits worth $250. Sponsored by Google Cloud. Learn more at

differences between mutect1 and mutect2 output

Hi I am trying to use mutect2 on my data.
Comparing mutect1.1.7 output and mutect2 output I noticed that the somatic calls are totally different.

Here is reported the mutect-1.1.7 command line:

java -jar /data01/Software/mutect/mutect-1.1.7.jar --analysis_type MuTect \
--reference_sequence human_hg19.fa \
--cosmic cosmic_chr.vcf \
--dbsnp dbSNP_144_00-All_chr.vcf \
--intervals SureSelect_All_Exon_V4_chr.list \
--input_file:tumor tumor.bam \
--input_file:normal normal.bam \
--out classical.out \
--coverage_file out.txt \
--vcf mutect1.vcf &

The input bam files are preprocessed using the GATK BaseRecalibrator and the GATK IndelRealigner.

Here is reported the mutect2 command line:

java -jar GenomeAnalysisTK_3.5.jar -T MuTect2 -R human_hg19.fa \
-I:tumor tumor.bam \
-I:normal normal.bam \
--dbsnp dbSNP_144_00-All_chr.vcf \
--cosmic cosmic_chr.vcf \
-L SureSelect_All_Exon_V4_chr.list \
-o mutect2.vcf

In this case the input bam files are preprocessed without the use of GATK IndelRealigner.

Comparing mutect1.vcf and mutect2.vcf there is only one common somatic variant call.


  • Geraldine_VdAuweraGeraldine_VdAuwera Cambridge, MAMember, Administrator, Broadie

    We do expect to see differences in the results between these two versions. MuTect2 has some dramatic operational changes compared to the original MuTect. Have you looked at their respective calls to evaluate whether the results look better in one case or the other? Can you describe the differences in more detail?

Sign In or Register to comment.