Hello,
I am using mutect2 in GATK v. 4.2.0 to call somatic mutations using 1 tumor and 2 normals for the same individual. I am a bit confused concerning the options and if Mutect behaves the same in both cases below. In the first below I provide 3 bams with 3 different ids in each and I specify the 2 ids as normal. In the second only the --tumor-sample is provided. Exclamation marks correspond to nextflow variables. What is the recommended command for calling somatic variants if you want to exclude information based on matched normals? I have the impression that when multiple normal sample ids are provided only one of them might be used but maybe I am wrong (because the second would require a second tumor sample).
Thanks!
gatk --java-options "-XX:-UsePerfData -XX:-UseParallelGC -Xmx16G" \
Mutect2 \
-L !{genome_parts_bed} \
-R !{ref_name} \
-O "result/!{tumor_sample_id}${PREFIX_FOR_OUTPUT}${part_prefix}.mutect2.vcf" \
--input !{tumor_sample_bam} \
--input !{second_sample_bam} \
--input !{third_sample_bam} \
--normal-sample !{second_sample_id} \
--normal-sample !{third_sample_id} \
--max-reads-per-alignment-start 90000 \
--bam-output "result/!{tumor_sample_id}${PREFIX_FOR_OUTPUT}${part_prefix}.realigned.bam" \
--bam-writer-type CALLED_HAPLOTYPES \
--native-pair-hmm-threads !{num_threads} !{extra_options}
gatk --java-options "-XX:-UsePerfData -XX:-UseParallelGC -Xmx16G" \
Mutect2 \
-L !{genome_parts_bed} \
-R !{ref_name} \
-O "result/!{tumor_sample_id}${PREFIX_FOR_OUTPUT}${part_prefix}.mutect2.vcf" \
--input !{tumor_sample_bam} \
--input !{second_sample_bam} \
--input !{third_sample_bam} \
--tumor-sample !{tumor_sample} \
--max-reads-per-alignment-start 90000 \
--bam-output "result/!{tumor_sample_id}${PREFIX_FOR_OUTPUT}${part_prefix}.realigned.bam" \
--bam-writer-type CALLED_HAPLOTYPES \
--native-pair-hmm-threads !{num_threads} !{extra_options}