{
  "id": 10010116,
  "title": "Pop-Corn: Predicting Perturbation Phenotype Effects Across Single-Cell and Spatial Contexts",
  "url": "https://urgent.news/2026/09/26/pop-corn-predicting-perturbation-phenotype-effects-across-single-cell",
  "topic": "science",
  "section": "Science",
  "published": "2026-09-26T00:00:00.000Z",
  "source": {
    "name": "bioRxiv",
    "slug": "biorxiv",
    "url": "https://www.biorxiv.org/content/10.64898/2026.09.25.754560v1?rss=1"
  },
  "original_language": "en",
  "account": "Perturbing genetic elements can modify the composition of cell populations, affecting the relative abundance of distinct cell types and states. Pooled single-cell screening techniques, like Perturb-seq, assess these responses across a broad range of perturbations. However, due to resource constraints, it is impractical to experimentally test all conceivable perturbations. The major challenge lies in predicting the compositional changes brought about by previously untested perturbations. Numerous perturbation-prediction approaches focus on forecasting gene-expression alterations and subsequently inferring cell-type and cell-state compositions. Surprisingly, even models that accurately predict gene-expression changes often struggle to accurately forecast these compositional shifts. To tackle this issue, researchers introduce Pop-Corn, a method that directly predicts how a perturbation reshapes cell-type composition without needing to reconstruct gene expression data. When applied to primary T-cell benchmarks, Pop-Corn outperformed expression-prediction pipelines in predicting the overall cell-state composition of unseen perturbations while also preserving the diversity of observed cell states. Moreover, Pop-Corn has been extended to intact tissues, enabling it to predict perturbation-induced changes in cell-type proportions within local cellular neighborhoods and generate hypotheses about context-dependent cellular interactions using attention patterns. Retrospective virtual screening results suggest that Pop-Corn can effectively prioritize perturbations for experimental follow-up based on their predicted effects on cell-state composition.",
  "summary": "Genetic perturbations can reshape cell populations by altering the relative abundance of specific cell types and states within the profiled population, including increases, decreases and states that become detectable after perturbation. Pooled single-cell screens, such as Perturb-seq, measure such responses at scale. However, only a small fraction of possible perturbations can be tested…",
  "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."
}