The abstract is read at the publisher; the summary is JClub's.
SCPM-based clustering using the Nankai GeneRank Classifier fine-tuned from Geneformer-V1 resolves colorectal cancer cell populations that classical methods cannot unambiguously identify.
- Why it matters: Training data with comprehensive cell type labels is limited, so new single-cell datasets can contain unrecognized populations that standard supervised approaches struggle to identify.
- What they did: The method fine-tunes Geneformer-V1 to create Nankai GeneRank Classifier, then clusters cells using softmax-normalized class probability matrix (SCPM) from its outputs.
- The result: On a public colorectal cancer dataset, the approach identifies previously ambiguous populations and enables discovery of novel candidate markers of colorectal cancer stem cells.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.