Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

SAVE: A Generalizable Framework for Multi-Condition Single-Cell Generation with Gene Block Attention

About

Modeling single-cell gene expression across diverse biological and technical conditions is crucial for characterizing cellular states and simulating unseen scenarios. Existing methods often treat genes as independent tokens, overlooking their high-level biological relationships and leading to poor performance. We introduce SAVE, a unified generative framework based on conditional Transformers for multi-condition single-cell modeling. SAVE leverages a coarse-grained representation by grouping semantically related genes into blocks, capturing higher-order dependencies among gene modules. A Flow Matching mechanism and condition-masking strategy further enhance flexible simulation and enable generalization to unseen condition combinations. We evaluate SAVE on a range of benchmarks, including conditional generation, batch effect correction, and perturbation prediction. SAVE consistently outperforms state-of-the-art methods in generation fidelity and extrapolative generalization, especially in low-resource or combinatorially held-out settings. Overall, SAVE offers a scalable and generalizable solution for modeling complex single-cell data, with broad utility in virtual cell synthesis and biological interpretation. Our code is publicly available at https://github.com/fdu-wangfeilab/sc-save

Jiahao Li, Jiayi Dong, Peng Ye, Xiaochi Zhou, Haohai Lu, Fei Wang• 2026

Related benchmarks

TaskDatasetResultRank
Perturbation predictionPBMC-IFN
CD8T PCC0.98
7
Batch effect correctionlung atlas
Bio. Score73
5
Batch effect correctionHeart
Bio. Score76
5
Batch effect correctionPBMC
Biological Conservation Score0.75
5
Dual condition generationHeart
WD8.3
4
Dual condition generationPBMC
WD5.37
4
Dual condition generationlung atlas
WD4.37
4
Gene level performance evaluationHeart
MSE0.01
4
Gene level performance evaluationPBMC
MSE0.02
4
Gene level performance evaluationLung
MSE0.01
4
Showing 10 of 20 rows

Other info

Follow for update