Single-Cell Model Hub
A curated collection of foundation models, virtual cells, and AI systems for single-cell biology — papers, code, and resources.
All Models
25
Contrastive Learning: 细胞界的"找不同"游戏：对比学习如何让AI在数十亿个细胞中自学成才
Contrastive learning approaches in single-cell models.
Chinese Teams: 在细胞的"语言"中寻找语法：中国团队的单细胞AI征途
Contributions from Chinese research teams in single-cell AI.
Multimodal: 一个细胞，三种语言：多模态大模型如何让AI读懂细胞的全部秘密
Multimodal single-cell models integrating multiple data types.
Hallucinations: 当 AI 把红细胞认成神经元：单细胞大模型的"幻觉"危机
Hallucination and uncertainty in single-cell AI models.
Batch Effects: 批次效应：单细胞大模型最大的敌人
Understanding and mitigating batch effects in single-cell data.
Cell2Sentence
SATURN (Species Alignment Through Unification of Rna and proteiNs)
scFoundation
Large-scale pre-trained model for single-cell transcriptomics with 100M parameters, trained on 50M+ cells.
scGPT
Generative pre-trained transformer for single-cell biology. Pre-trained on 33M+ cells from CELLxGENE.
Virtual Cell Models
Generative models for simulating single-cell responses to perturbations.
UCE
Universal Cell Embedding — species-agnostic cell representation model for cross-species analysis.
CELLxGENE Census Models
Foundation models from CZI, trained on CELLxGENE Census data (50M+ cells).
Geneformer
Context-aware foundation model for gene network predictions. Transfer learning from 30M single-cell transcriptomes.
scBERT
BERT-based pre-trained model for single-cell transcriptomics with gene-level tokenization.
TranscriptFormer
TranscriptFormer 用 5.4 亿参数、1.12 亿个细胞、12 个物种，在 one transformer 里读懂了从酵母到人类的"细胞通用语法"——跨物种细胞分类、零样本疾病识别、系统发育树涌现，单细胞大模型正式从"全基因组时代"跨入了"全演化树时代"
