Mola: a sequence-to-distribution model predicts single-molecule chromatin states
anshulkundaje · x · 2026-11-06
A University of Chicago team led by Jian Zhou released a bioRxiv preprint introducing Mola, a sequence-to-distribution (S2d) model. Existing deep learning models predict cell-population averages and miss single-molecule heterogeneity; Mola instead models the full probabilistic distribution of single-molecule chromatin accessibility states directly from DNA sequence across 10 kb regions. Using Mola, the team quantified sequence-driven nucleosome regulation and uncovered distinct modes of transcription factor engagement with nucleosomes. Concurrent studies include SMFNet (Anshul Kundaje's group) and ChAI-Net (Jodi Lee, Katie Pollard, Vijai Ramani).
Related event: Mola Models Single-Molecule Chromatin State Distributions from DNA Sequence(4 posts)→
More from Research
- Study: Epigenetic aging accelerated in kids and adults during the Covid pandemic — EricTopol · 2026-11-06
- Memorilla adds latent semantic memory to LLMs, positioned beyond RAG — StanfordAILab · 2026-11-06
- Team finds models trained for collaboration match frontier coding without explicit coding training — gharik · 2026-11-06
- NVIDIA open-sources Dory1.0-4B: looped latent-recurrent LLM beats Qwen3.5-9B on 12 benchmarks — ChengleiSi · 2026-11-06
- Rohit Krishnan maps "AGI institutions" experiments: 11 strong matches, zero at national scale — soumitrashukla9 · 2026-11-06
- Anneal V1 Core released: CPU-native low-parameter state-space architecture under CC0 — True-Beach1906 · 2026-11-06