Skip to content
Open access

A knowledge-driven framework for predicting single-cell responses for unprofiled drugs

Aug 2026 · Nature Machine Intelligence · Vol 8, pp. 1478 - 1491 · 2 citations · 52 references

TL;DR

An artificial intelligence framework that integrates biological mechanism knowledge to predict how cells respond to chemical perturbation, improving generalization to untested drugs and prioritizing cancer drug candidates in virtual screening is introduced.

Abstract

Predicting cellular response to chemical perturbations is critical to build virtual cells, yet experimentally profiled compounds cover only a small fraction of this space. Existing models struggle to generalize to unprofiled compounds, as they typically treat drugs as isolated identifiers without encoding their mechanistic relationships. Here, we present MAP, a framework that integrates structured biological knowledge into cellular perturbation modelling and supports zero-shot prediction for small molecules with scarce or absent profiles. (1) We construct MAP-KG, a knowledge graph that unifies 14 public resources, spanning 187,089 drugs, 22,924 genes and 694,246 mechanistic relationships. (2) We propose a knowledge-driven pretraining strategy that aligns molecular structures, protein sequences and textual mechanistic descriptions into a unified embedding space, producing mechanism-aware and transferable gene and compound embeddings. These representations are then coupled with a pretrained single-cell foundation model to condition perturbation response prediction. (3) We evaluate MAP under two zero-shot generalization regimes: unseen cell type–drug combinations and a stricter setting of unprofiled drugs, where it improves the top-50 differentially expressed gene Pearson delta correlation by up to +12.3% and +11.8%, respectively, over the strongest baselines across three benchmarks. We further perform pathway-level functional analysis via gene set enrichment analysis for in silico screening, where MAP predicts mechanism-consistent programmes on unprofiled candidate drugs, and prioritizes four out of five approved anti-cancer drugs in A-549 (non-small-cell lung cancer). Feng et al. introduce MAP, an artificial intelligence framework that integrates biological mechanism knowledge to predict how cells respond to chemical perturbation, improving generalization to untested drugs and prioritizing cancer drug candidates in virtual screening.

Read PDF

Similar papers

Open access Aug 2026

A pretrained unified model enables cellular functional profile prediction and multi-objective virtual drug screening

Cells are characterized by molecular states, coordinated molecular interactions, regulatory programs, and responses to perturbations. Systematic mapping of these cellular functional profiles across biological contexts remains experimentally costly and fragmented. Here we present InsilicoCell, a pretrained multi-modal,...

Ruoqiao Chen, Li Huang, Yu Qiao et al. · 0 citations
Open access Sep 2026

Regulon-informed cellular representations reveal task-dependent generalization in drug combination prediction

A symmetric three-class predictor of DrugComb ZIP interactions was developed and compared all seven combinations of landmark expression, Hallmark pathway scores and CollecTRI regulon activities, demonstrating that the predictive value of functional context representations depends on both the type of domain shift and th...

Elizaveta Ignatova, M. Likhter · 0 citations
Open access Aug 2026

A mechanism-annotated benchmark reveals limited fidelity to drug-response signatures in single-cell perturbation models

Single-cell drug perturbation models are increasingly used to predict how compounds remodel cellular states, but they are still largely assessed by expression reconstruction. Whether high expression similarity reflects preservation of drug-response signatures remains unclear. Here we present scDrugPerturb-Bench, a mech...

Le-Hang Li, Shaoming Duan, Xin-Xiang Zha et al. · 0 citations
Sep 2026

An operational perturbation proteomics-based virtual cell model.

This work generated more than 38 million temporal protein-abundance measurements from systematically perturbed breast cancer cell lines, and developed ProteinTalks, a virtual cell model that achieves higher performance than the selected benchmark implementations under the evaluated protocols.

R. Sun, Liujia Qian, Yong-Ge Li et al. · 1 citation
#diffusion models Open access Sep 2026

scLDM: a conditional diffusion framework for single-cell perturbation prediction

Overall, scLDM provides a robust and biologically consistent strategy for in silico perturbation screening, and exhibits strong interpretability, as the learned perturbation embeddings show high functional alignment with known biological mechanisms.

Bo-Yang Wu, Yu-Hang Liu, Yue Cheng et al. · 0 citations
Open access Sep 2026

A biological-response compound representation allows chemical perturbation prediction across cell lines

Accurately predicting cellular responses to drugs remains a challenge with the potential to reduce experimental screening costs and accelerate drug discovery. Current computational approaches represent compounds through chemical structures, which carry little information regarding their activity within biological syste...

Léa Kaufmann, Lola Le Breton, E. Carraz-Billat et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.