This repository serves as a comprehensive, curated collection of resources for researchers in the rapidly evolving field of AI-powered virtual cell research. It aims to consolidate papers, datasets, benchmarks, talks, and community initiatives, providing a centralized hub for understanding, modeling, and building virtual cells and related cellular foundation models. The target audience includes AI researchers, computational biologists, and anyone seeking to advance the state of cellular simulation and prediction.

How It Works

The "Awesome Virtual Cell" repository functions as a curated directory, meticulously gathering and organizing resources relevant to AI-driven virtual cell research. It adheres to specific inclusion rules, prioritizing peer-reviewed publications, high-signal preprints, and official project pages. The collection spans a broad yet practical scope, encompassing virtual cell perspectives, perturbation modeling, single-cell and multimodal foundation models, spatial and morphology modeling, and biological AI agents.

Quick Start & Requirements

This repository is a curated list of resources, not a deployable software project. There are no installation or execution commands. Users can navigate the content by browsing the organized sections within the README, such as "Overview Papers," "Research Papers," and "Datasets."

Highlighted Details

• Extensive Resource Coverage: Features a vast array of peer-reviewed papers from top journals (Nature, Cell, Science), numerous datasets covering perturbation atlases, single-cell references, multimodal data, morphology, and spatial omics, alongside reports, blogs, and videos.
• Focus on Cutting-Edge Research: A significant portion of the listed resources are from 2025 and 2026, highlighting the repository's commitment to showcasing the latest advancements in AI for virtual cell biology.
• Key Research Areas: Emphasizes foundational models for cellular biology, perturbation modeling, spatial transcriptomics, AI agents for biological discovery, and the development of robust benchmarking methodologies.
• Community Engagement: Includes links to community challenges and initiatives like the "Virtual Cell Challenge," fostering collaborative development and evaluation in the field.

Maintenance & Community

Contributions are welcomed via GitHub Issues or Pull Requests, following the guidelines in CONTRIBUTING.md. The repository highlights community efforts such as the "Virtual Cell Challenge" and provides links to related projects and software environments like the "Arc Virtual Cell Atlas" and "VCell Software."

Licensing & Compatibility

The repository itself does not specify a license. The linked resources (papers, datasets, software) will have their own respective licenses and usage terms, which users must ascertain individually. Compatibility for commercial use or closed-source linking depends entirely on the licenses of the individual resources cited.

Limitations & Caveats

This is a curated list, and its value is contingent on the ongoing maintenance and accuracy of the links and descriptions. Generic biomedical AI work with weak cell-modeling relevance, low-confidence sources, or broken links are intentionally excluded but may still arise. Users must independently verify the applicability and licensing of each cited resource.