{
  "id": 18107857,
  "title": "Perturbation-induced gene programs map the hidden regulatory architecture of complex diseases",
  "url": "https://urgent.news/2026/11/04/perturbation-induced-gene-programs-map-the-hidden-regulatory",
  "topic": "science",
  "section": "Science",
  "published": "2026-11-04T00:00:00.000Z",
  "source": {
    "name": "bioRxiv",
    "slug": "biorxiv",
    "url": "https://www.biorxiv.org/content/10.64898/2026.11.03.762080v1?rss=1"
  },
  "original_language": "en",
  "account": "The study introduces Perturbscape, a computational framework designed to uncover multiscale gene programs triggered by individual perturbations within CRISPR Perturb-seq screens. By employing both linear and nonlinear contrastive methods, as well as supplementary program-learning techniques, Perturbscape integrates this data with human genetic information to identify disease-relevant genes, perturbations, and downstream regulatory pathways.\n\nWhen analyzing 15,859 perturbation-by-context profiles from seven Perturb-seq systems and 53 complex diseases and traits, the researchers found that perturbation-induced programs demonstrated up to 4.3-fold higher heritability informativeness compared to traditional steady-state multi-tissue gene prioritization. Moreover, these programs maintained significant conditional heritability informativeness across 34 cell system-trait comparisons, even after adjusting for steady-state prioritization methods.\n\nFurthermore, Perturbscape prioritization yielded 2.5-5.3-fold higher trait relevance scores than gene prioritization derived from control cells, pooled perturbations, or differential-expression summaries. In the K562 cell line, genes identified as significant by Perturbscape exhibited stronger enrichment for rare-variant burden genes than those prioritized using traditional methods. Additionally, these Perturbscape-selected genes were corroborated by independent enhancer CRISPRi screens.\n\nIn the HepG2 cell line, perturbation-induced programs revealed statin-responsive pathways that were not significantly enriched by steady-state PoPS (Predictive Oncogenic Signature) prioritization, supported by both pharmacological transcriptional responses and population plasma proteomics analysis. Perturbscape also uncovered blood-related trait-conditioned trans-regulatory relationships, which were specifically supported by population trans-eQTLs (transcriptome-encoded quantitative trait loci) in IFN-gamma-stimulated monocytes (OR = 1.30, P=1.6e-05).\n\nAdditionally, Perturbscape uncovered developmentally stage-specific convergence between regulators of monogenic diabetes and polygenic type 2 diabetes risk during pancreatic islet differentiation. Finally, the framework revealed class-specific neuronal regulatory programs in vivo associated with autism. These findings collectively demonstrate that perturbation-induced gene programs serve as a valuable complementary layer of disease-relevant regulatory information, uncovering genetically informative pathways and context-specific regulatory relationships that are often underrepresented in steady-state molecular atlases.",
  "summary": "Steady-state molecular profiles incompletely capture the regulatory programs through which genetic variation influences complex disease, particularly programs that emerge only when cells are perturbed. Here, we introduce Perturbscape, a computational framework that learns multiscale gene programs for individual perturbations in a CRISPR Perturb-seq screen using linear and nonlinear contrastive…",
  "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."
}