Article

Single-cell foundation models reveal context-sensitive cancer programmes under subtype shift

2026-05-01

Abstract excerpt

Single-cell foundation models (scFMs) have shown promise as transferable representations of cellular state, but recent zero-shot evaluations suggest that they do not consistently outperform simpler baselines. We asked whether this apparent limitation reflects an intrinsic weakness of scFMs or instead the difficulty of using them without task-specific adaptation. To test this, we fine-tuned two widely used scFMs, G...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: 767e034d-3ac4-5a7f-b74a-7537d26abdee
DOI: 10.64898/2026.04.28.721114

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Single-cell foundation models reveal context-sensitive cancer programmes under subtype shiftDOI 10.64898/2026.04.28.721114Read

Quantized multi-task learning for context-specific representations of gene network dynamicsSemantic proximity Read

Cross-Tissue Generalization of Rare Cell-Type Representations in Single-Cell Foundation Models: A Benchmark Using Public Reference AtlasesSemantic proximity Read

Gene-Family Encoding Boosts Domain-Adapted Single-Cell Language ModelsSemantic proximity Read

GeneBag: training a cell foundation model for broad-spectrum cancer diagnosis and prognosis with bulk RNA-seq dataSemantic proximity Read

CancerFoundation: A single-cell RNA sequencing foundation model to decipher drug resistance in cancerSemantic proximity Read

GeneBag: training a cell foundation model for broad-spectrum cancer diagnosis and prognosis with bulk RNA-seq dataSemantic proximity Read

evoCancerGPT: Generating Zero-Shot Single-Cell and Single-Sample Cancer Progression Through Transfer LearningSemantic proximity Read

Prototype-based continual cell-type annotation reveals cellular state transitions in expanding single-cell atlasesSemantic proximity Read

When Simpler Models Win: A Calibration Benchmark of scGPT for Cell-Type AnnotationSemantic proximity Read

scOPE identifies which driver-associated expression programs transfer from bulk tumors to single cellsSemantic proximity Read

scLong: A Billion-Parameter Foundation Model for Capturing Long-Range Gene Context in Single-Cell TranscriptomicsSemantic proximity Read

Tahoe-x1: Scaling Perturbation-Trained Single-Cell Foundation Models to 3 Billion ParametersSemantic proximity Read