scBaseCount: An AI agent-curated, standardized, auto-updated single-cell data repository

scBaseCount is now the largest public single-cell RNA-seq repository, encompassing over 502 million cells from 27 organisms and 75 tissues, all uniformly reprocessed from 10× Genomics datasets in the SRA. An autonomous AI agent discovers, annotates, and harmonizes new datasets continuously, eliminating batch effects that hamper cross-study comparisons. The resource is designed to support both basic cell-biology research and the training of foundation AI models, with direct applicability to tumor microenvironment and cancer cell-state analyses.