AI Insight
Researchers developed COMPASS, a computational model that predicts how genetic perturbations affect gene expression in cells by decomposing responses into shared and gene-specific components. Analyzing 2,270 CRISPR interference perturbations across six cell lines, they found that perturbation responses exist on a continuum from broadly shared effects to targeted gene-specific changes, with a perturbation's position on this continuum being conserved across different cell lines. COMPASS outperformed existing models including scGPT and GEARS in both accurately predicting transcriptional responses and distinguishing between different perturbations.
Why it matters
This work could improve our ability to predict cellular responses to genetic interventions, which is valuable for drug development, understanding disease mechanisms, and designing targeted genetic therapies. The finding that perturbation effects are partially conserved across cell types suggests responses may be more predictable than previously thought.
Understand the Science
⚠️ Preprint – Noch nicht peer-reviewed
Dieser Artikel wurde noch nicht von unabhängigen Experten begutachtet. Die Ergebnisse sind vorläufig und sollten mit Vorsicht interpretiert werden.
Predicting how a genetic perturbation reshapes a cell’s transcriptome is a central goal of computational biology. Previous studies report that the mean response across training perturbations rivals specialized models on standard accuracy metrics, even though it cannot distinguish which perturbation occurred. Across 2,270 CRISPRi perturbations measured in each of six cell lines, we show that this apparent paradox reflects a conserved organization of perturbation responses. Perturbations span a continuum from responses strongly aligned with the mean to more targeted responses that depart from it. Crucially, a perturbation’s position along this continuum is conserved across cell lines (Kendall’s W=0.59) and predictable from STRING protein-interaction embeddings (R^2=0.35). We formalize this structure with COMPASS, an interpretable linear model that decomposes each response into shared and gene-specific components and estimates them separately. The shared-response component is modeled as a cell-line-wide response scaled by a perturbation-specific coefficient. This coefficient is strongly conserved across cell lines. The residual gene-specific component—which is moderately conserved across cell lines—recovers pathway-level programs. COMPASS outperforms scGPT, CPA, GEARS, GenePert, and State in both response accuracy (de-biased Pearson delta 0.34 vs. <=0.32) and perturbation discrimination (cosine PDS gain 0.23 vs. <=0.08). These results recast perturbation prediction across cellular contexts as component-wise inference, with each component estimated from the evidence best suited to it.
Source: COMPASS: Component-Wise Inference of Shared and Gene-Specific Perturbation Response