I was trying to run the Mutect2 pipeline through the Sarek Nextflow pipeline (https://github.com/nf-core/sarek) and encountered an issue with the GatherPileupSummaries command used. As I am trying to do WES somatic calling, multiple pileup tables were generated for the various intervals used. The pipeline then tries to merge the pileup tables using GatherPileupSummaries, and feeds the merged table into CalculateContamination. It turns out that the merged table is not actually sorted by variants, which leads to CalculateContamination attempting to create incorrect intervals depending on what order the original files are in. The exact commands and outputs are documented here: https://github.com/nf-core/sarek/issues/899. While this is not really an urgent bug, I think it would still be desirable for GatherPileupSummaries to enforce some sorting on the merged pileup table to prevent such problems from occurring.