mofaplus-multi-omics

What problem does it solve?

This Skill enables the joint decomposition of multiple omics layers (e.g., scRNA-seq, ATAC-seq, proteomics) into a reduced set of latent factors, revealing underlying biological variation and relationships across modalities.

Core Features & Use Cases

• Multi-Omics Integration: Combine data from different biological measurements (RNA, ATAC, protein, methylation) from the same samples.
• Latent Factor Discovery: Identify hidden biological drivers and patterns that span across omics types.
• Dimensionality Reduction: Reduce complex multi-omics data into a more manageable set of interpretable factors.
• Use Case: Integrate single-cell RNA sequencing and ATAC sequencing data to discover cell states defined by coordinated gene expression and chromatin accessibility patterns.

Quick Start

Use the mofaplus-multi-omics skill to train a MOFA+ model on your RNA and ATAC AnnData objects, saving the output to 'mofa_model.hdf5'.