Mechanistic gene-regulation-based trajectory inference with bowside

bowside, a shallow neural stochastic differentiation model, recovers gene expression trends and underlying regulatory interactions from single-cell RNA sequencing, including known fate-determining network motifs.

• Why it matters: Single-cell RNA sequencing can profile differentiation, but jointly inferring trajectories and reconstructing gene regulatory networks is still difficult.
• What they did: The approach mechanistically models gene regulation using a shallow neural stochastic differentiation framework and is applied to hematopoiesis and induced pluripotent stem cell reprogramming datasets.
• The result: In both applications, it retrieves known fate-determining network motifs, enabling simultaneous trajectory inference and gene regulatory network reconstruction.

AI summary of the abstract · may contain errors · not medical advice · read the paper.