mmVelo: A deep generative model for estimating cell state-dependent dynamics across multiple modalities.

The article discusses the development of a deep generative model called mmVelo, which aims to estimate cell state-dependent dynamics across multiple modalities in single-cell multiomics data. By analyzing spliced and unspliced mRNA expression, mmVelo projects dynamics onto chromatin accessibility, enhancing our understanding of regulatory relationships across molecular layers. The model was applied to data from a developing mouse brain and validated for accuracy, identifying crucial transcription factors for chromatin accessibility regulation in mouse skin. Additionally, the study demonstrates the ability to infer missing modalities during human brain development through cross-modal generation. [Extracted from the article]