GEOapplication/xmlftp://ftp.ncbi.nlm.nih.gov/geo/series/GSE295nnn/GSE295214/primaryOK200TranscriptomicsHomo sapiensExpression profiling by high throughput sequencinghttps://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE295214GEOGSEfalseDeciphering molecular underpinnings of cell state and state transitions with a global cell-state manifoldMammalian organisms comprise numerous types of cells with distinct gene-expression states, which arise from single zygotes differentiating on a complex landscape of cell states. The surge of single-cell transcriptomic data from human and other animals offer the possibility of constructing a global cell-state map, which will be foundational for understanding the molecular basis of normal tissue functions and dysfunction in diseases. However, this challenging task requires computationally efficient integration of many large datasets and accurate removal of artifactual batch effects while preserving true biological variations. Here, we developed a contrastive learning model that combines the strong batch-correction power of pairwise integration with the computational scalability of deep learning and used this model to construct a global cell-state map from human and mouse single-cell transcriptomic data. This model outperforms other state-of-the-art foundation models for batch correction and dataset integration. The resulting global cell-state manifold enables projection and annotation of cell states, generation of cell-state trajectories, and prediction of causal genes driving cell-state transitions. To further illustrate its power for understanding the molecular basis of cell-state transitions, we performed genome-scale perturb-seq experiments on human embryonic stem cells and systematically mapped the cell states induced by genetic perturbations to the global cell-state manifold. We identified gene programs that are differentially regulated both under normal development and upon genetic perturbations and distinct classes of perturbation-induced cell-state transitions that are aligned with or orthogonal to normal cell differentiation. The global cell-state manifold and related computational tools provide a powerful resource for understanding cell states and transitions in health and disease.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 sapiens