AI virtual cell modelsAI 虚拟细胞模型
Review pending待审阅 Version 0.1.0 · status: draft · last expert review: not yet reviewed版本 0.1.0 · 状态：初稿 · 最近专家审阅：尚未审阅
Research question研究问题
- What is an AI virtual cell model?什么是 AI 虚拟细胞模型？
- Which virtual cell models have been applied to glioblastoma?哪些虚拟细胞模型已应用于胶质母细胞瘤？
- How reliable are current virtual cell predictions?当前虚拟细胞的预测有多可靠？
Current understanding当前认识
An AI virtual cell is a model intended to simulate how cell states change under perturbations. Many general models exist, but GBM-specific applications are few and early, and benchmarks show that perturbation predictions often do not beat simple baselines.AI 虚拟细胞是旨在模拟细胞状态在扰动下如何变化的模型。通用模型已有不少，但 GBM 特异的应用很少且处于早期；基准研究显示扰动预测常常不能超越简单基线。
Bunne et al. (Cell, 2024) define the AI virtual cell as built on multimodal, multiscale, universal cell representations that simulate in silico how cell states change under perturbations; a “world model” framing has since been published.Bunne 等（Cell，2024）将 AI 虚拟细胞定义为基于多模态、多尺度通用细胞表征、在计算机中模拟细胞状态在扰动下变化的模型；此后又发表了“世界模型”框架。 VC0001 VC0131
Building blocks include single-cell foundation models (scGPT, Geneformer, scFoundation) and perturbation predictors (GEARS, CPA, Arc Institute’s State).其组成部分包括单细胞基础模型（scGPT、Geneformer、scFoundation）与扰动预测模型（GEARS、CPA、Arc 研究所的 State）。 VC0008 VC0009 VC0011 VC0013 VC0014 VC0019
GBM-specific applications in the library: scFOCAL/ISOSCELES (state–drug signature integration, peer reviewed), DELPHAI (temozolomide in organoids, preprint) and a proteome-calibrated mechanistic digital twin of the GBM astrocyte (preprint).文献库中的 GBM 特异应用：scFOCAL/ISOSCELES（状态-药物特征谱整合，已同行评审）、DELPHAI（类器官替莫唑胺，预印本）及蛋白组校准的 GBM 星形细胞机制性数字孪生（预印本）。 VC0132 VC0129 · preprint · 预印本 VC0071 · preprint · 预印本
Reliability: independent benchmarks report that deep perturbation models do not yet reliably outperform simple linear baselines, zero-shot foundation-model performance is limited, and systematic variation can inflate apparent accuracy. A preprint (X-Cell) claims to surpass linear baselines; it is not yet peer reviewed.可靠性：独立基准研究报告深度扰动模型尚不能稳定超越简单线性基线，基础模型零样本表现有限，系统性变异会夸大表观准确度。X-Cell 预印本声称超越线性基线，但尚未经同行评审。 VC0048 VC0050 VC0052 VC0033 · preprint · 预印本
Evidence map证据图谱
Primary studies & datasets原始研究与数据集
- VC0001 How to build the virtual cell with artificial intelligence: Priorities and opportunities. — Bunne C et al., Cell, 2024
- VC0131 A world model of the virtual cell — Xing et al., Cell, 2026
- VC0019 Predicting cellular responses to perturbation across diverse contexts with State. — Adduri AK et al., Cell, 2026
- VC0008 scGPT: toward building a foundation model for single-cell multi-omics using generative AI. — Cui H, Wang C, Maan H, Pang K, Luo F, Duan N, Wang B., Nature methods, 2024
- VC0009 Transfer learning enables predictions in network biology. — Theodoris CV, Xiao L, Chopra A, Chaffin MD, Al Sayed ZR, Hill MC, Mantineo H, Brydon EM, Zeng Z, Liu XS, Ellinor PT., Nature, 2023
- VC0011 Large-scale foundation model on single-cell transcriptomics. — Hao M, Gong J, Zeng X, Liu C, Guo Y, Cheng X, Wang T, Ma J, Zhang X, Song L., Nature methods, 2024
- VC0013 Predicting transcriptional outcomes of novel multigene perturbations with GEARS. — Roohani Y, Huang K, Leskovec J., Nature biotechnology, 2023
- VC0014 Predicting cellular responses to complex perturbations in high-throughput screens. — Lotfollahi M et al., Molecular systems biology, 2023
Related models & tools相关模型与工具
- VC0132 Drug and single-cell gene expression integration identifies sensitive and resistant glioblastoma cell populations — Suter et al., Nature Communications, 2026
- VC0129 · preprint · 预印本 DELPHAI, AI Agent for Predicting Drug Response and Resistance — Peng et al., preprint, 2025
- VC0071 · preprint · 预印本 A Proteome-Calibrated Mechanistic Digital Twin of the Glioblastoma Astrocyte Derived from Patient Multi-Omic Data — Mayfield JD., bioRxiv, 2026
- VC0033 · preprint · 预印本 X-Cell: Scaling Causal Perturbation Prediction Across Diverse Cellular Contexts via Diffusion Language Models — Wang C et al., bioRxiv, 2026
Perturbation & validation evidence扰动与验证证据
- VC0048 Deep-learning-based gene perturbation effect prediction does not yet outperform simple linear baselines. — Ahlmann-Eltze C, Huber W, Anders S., Nature methods, 2025
- VC0049 Benchmarking foundation cell models for post-perturbation RNA-seq prediction. — Csendes G, Sanz G, Szalay KZ, Szalai B., BMC genomics, 2025
- VC0050 Zero-shot evaluation reveals limitations of single-cell foundation models. — Kedzierska KZ, Crawford L, Amini AP, Lu AX., Genome biology, 2025
- VC0052 Systema: a framework for evaluating genetic perturbation response prediction beyond systematic variation. — Viñas Torné R, Wiatrak M, Piran Z, Fan S, Jiang L, Teichmann SA, Nitzan M, Brbić M., Nature biotechnology, 2025
- VC0002 Virtual Cell Challenge: Toward a Turing test for the virtual cell. — Roohani YH et al., Cell, 2025
- VC0058 Virtual Cell Challenge 2026: Benchmarking zero-shot generalization across cellular contexts. — Arc Virtual Cell Initiative Team, Goodarzi H., Cell, 2026
19 library items cited on this page; 3 are preprints (not peer reviewed).本页引用文献库条目 19 条，其中预印本（未经同行评审）3 条。
What remains uncertain尚不确定之处
Most training perturbation data come from cell lines such as K562 and RPE1, not brain tumor cells.大部分扰动训练数据来自 K562、RPE1 等细胞系，而非脑肿瘤细胞。 VC0040
Generalisation to unseen cellular contexts is exactly what the 2026 Virtual Cell Challenge tests; results are pending.向未见细胞情境的泛化正是 2026 虚拟细胞挑战赛的考察内容，结果尚待公布。 VC0058
Two of three GBM-specific model items are preprints.三个 GBM 特异模型条目中有两个是预印本。 VC0129 · preprint · 预印本 VC0071 · preprint · 预印本
For research and education only. Not medical advice and not a recommendation of any diagnosis, treatment or product. Patients should consult their own physicians.仅供科研与教育参考，不构成医疗建议，也不构成对任何诊疗或产品的推荐。患者请咨询主治医生。