A Pan-Cancer Single-Cell Atlas to Evaluate Tumor Identity, Cell Line Concordance, and Dependency Mapping

Single-cell RNA sequencing (scRNA-seq) overcomes these limitations, yet biological interpretability of public datasets is often compromised by variable data quality, inconsistent annotation, and atlas-scale aggregation strategies that favor data volume over biological coherence.

We therefore developed a stringent integration framework that prioritizes representative malignant transcriptional states. Using Mahalanobis distance-based selection in batch-corrected latent space, we constructed a pan-cancer atlas of 135,441 high-quality malignant cells from 494 samples spanning 36 adult and pediatric cancer types. Atlas-derived signatures were used to assess tumor–cell line concordance and project ElasticNet models trained on DepMap CRISPR screens to infer cancer-specific gene dependencies.