This is the orchestrator for working with multiple omics layers at once. It handles the messy parts of integrating RNA-seq, proteomics, methylation, genomics, and metabolomics data: matching samples across datasets, mapping features to common gene identifiers, running cross-omics correlations (like checking if high mRNA actually translates to high protein), and doing multi-omics clustering with methods like MOFA+ or SNF. You'd reach for this when you have patient data across multiple molecular layers and need to find concordant signals, identify regulatory chains (SNP to methylation to expression), or do patient subtyping with combined evidence. The documentation is unusually honest about discordance being interesting biology rather than just noise.
npx skills add https://github.com/mims-harvard/tooluniverse --skill tooluniverse-multi-omics-integration