SMFnet reveals principles of combinatorial transcription factor assembly

SMFNet predicts distributions of single-molecule chromatin accessibility profiles from DNA sequence at cis-regulatory elements, linking distinct chromatin states to specific TF binding motifs.

• Why it matters: Population-averaged accessibility models obscure distinct molecular chromatin states that may correspond to different TF assemblies, limiting how individual TFs and combinations shape chromatin opening.
• What they did: The study introduces SMFNet, a generative diffusion model, and PRISM to identify sequence features tied to individual molecular states, then uses motif perturbations and controlled in silico motif screens.
• The result: Motif perturbations and in silico screens redistribute and dissect TF-dependent chromatin-opening signatures, and genetic perturbation via natural variation at endogenous cRE confirms TF-specific effects on accessibility in vivo.

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