Single-Cell Perturbation Modeling

Ask about this topic

Overview

Single-cell perturbation modeling asks how gene edits, drugs, and other interventions change cell states, including how responses differ across cells and conditions. Approaches include conditional flow-matching and diffusion models that generate response distributions, latent-variable models that separate perturbation effects from baseline cell state, and methods using regulatory networks, biological knowledge, or retrieval of similar experiments. Models are tested on settings such as Perturb-seq, drug-response atlases, and the Virtual Cell Challenge, with tasks including unseen perturbations, cell types, doses, and combinations. A central challenge is that strong benchmark scores may not transfer to new contexts or show that models learned target-specific mechanisms rather than shared background signals.

Latest papers

CardsList

61 papers9 papers in the last four weeks, up 13% on the four weeks before. 0.1% of all new papers.

Aug 4, 2025

AI Scientists for Building Virtual-Cell Models

Autonomous agents, from virtual laboratories to biological data-analysis agents and co-scientists, have begun to turn scientific goals into hypotheses. Whether they can go further and convert real, noisy and complex biological data into executable model hypotheses and architectures that encode testable assumptions remains unclear. Predicting how cells respond to genetic and chemical perturbations is a natural test of this ability: it is central to decoding disease mechanisms and therapeutic discovery, yet it demands cross-disciplinary expertise rarely concentrated in a single research group. Here, we introduce CellForge, an organized team of autonomous agents for single-cell perturbation modeling. Given a dataset and a task, it surveys the literature, parses the data, identifies relevant baselines, and formulates a model hypothesis through critic-mediated discussion among agents specialized in data modeling, single-cell biology, and deep learning. It then writes the training code, runs the experiments, and returns a trained predictor including architecture, code, and validation results. The architecture itself is the hypothesis: CellForge produces it end-to-end, with no human specifying the design, loss, or training protocol. Across six perturbation datasets spanning transcriptomic, epigenomic, and proteomic contexts, CellForge yields models that are competitive with hand-crafted state-of-the-art results on established tasks and extends the same workflow to paired multi-omic readouts. Blinded human evaluators rate CellForge's research plans as more scientifically sound and innovative than those from single LLMs or deep-research systems. In summary, we show autonomous agents can complete the full hypothesis-to-validation cycle in single-cell biology, with an AI effectively building another AI, in a complex setting that currently requires a team of human experts.

Xiangru Tang, Jiapeng Chen, Daniel Shao +12

Yale University · University of Pennsylvania · Helmholtz Zentrum München +3

AI Agents for Scientific DiscoveryAutonomous Research AgentsSingle-Cell Perturbation ModelingLLM Agent Orchestration
AskPDF