The abstract is read at the publisher; the summary is JClub's.
scGSL, a graph neural network for tumor scRNA-seq, achieves 84.83% average accuracy (F1 80.92%) for cross-patient cell type annotation across 49,024 cells from 19 patients.
- Why it matters: Tumor microenvironment heterogeneity is hard to interpret because scRNA-seq lacks spatial context and ligand-receptor interaction data are incomplete, limiting accurate cell type labeling and cell-cell communication inference.
- What they did: Using 49,024 cells from 19 patients spanning leukemia, breast invasive carcinoma, and colorectal cancer, the model performs graph-based transfer learning for cell type prediction and unsupervised gene interaction discovery.
- The result: scGSL identifies biologically meaningful gene interactions in an unsupervised manner, supported by significant expression differences in key gene pairs across cancers, alongside improved annotation metrics.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.