{
  "id": 3336253,
  "title": "MultiFlow: coupled flow matching for predicting single-cell multiomic perturbation responses in unseen cellular contexts",
  "url": "https://urgent.news/2026/08/25/multiflow-coupled-flow-matching-for-predicting-single-cell-multiomic",
  "topic": "science",
  "section": "Science",
  "published": "2026-08-25T00:00:00.000Z",
  "source": {
    "name": "bioRxiv",
    "slug": "biorxiv",
    "url": "https://www.biorxiv.org/content/10.64898/2026.08.20.746112v1?rss=1"
  },
  "original_language": "en",
  "account": "MultiFlow, a new framework for predicting cellular responses to perturbation, has been developed by researchers. This coupled flow-matching method brings together the generation and prediction of paired gene expression and chromatin accessibility data. By learning flows that condition on perturbation and control-derived cellular-state representations, MultiFlow can predict coordinated multiomic responses in cellular contexts that have not been seen before.\n\nIn tests comparing multiomic generation benchmarks, MultiFlow was able to accurately reproduce paired RNA-ATAC states and their population distributions. When it came to multiomic perturbation benchmarks, MultiFlow demonstrated the strongest overall performance in predicting both gene expression and chromatin accessibility responses. This is an improvement over previous methods that only focused on individual molecular layers.\n\nThe researchers found that the joint modeling of multiple molecular layers in MultiFlow helped preserve the coordination of perturbation-induced changes in both RNA-ATAC states. This includes maintaining the relationship between peaks and genes, as well as the structure of cellular neighborhoods across different molecular layers. These findings demonstrate that coupled flow matching can serve as a unified generative framework for modeling paired multiomic states and predicting coordinated perturbation responses across various cellular contexts. The code and tutorial for implementing MultiFlow are available online for those interested in utilizing this new approach.",
  "summary": "Predicting cellular responses to perturbation requires resolving coordinated changes across molecular layers, yet most single-cell perturbation models focus on transcriptional responses alone. Here we present MultiFlow, a coupled flow-matching framework that unifies generation and perturbation prediction of paired gene expression and chromatin accessibility. By learning coupled RNA-ATAC flows…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}