Article

Sparse autoencoders reveal organized biological knowledge but minimal regulatory logic in single-cell foundation models: a comparative atlas of Geneformer and scGPT

2026-03-25

Abstract excerpt

<title>Abstract</title> <p>Background: Single-cell foundation models such as Geneformer and scGPT encode rich biological information, but whether this includes causal regulatory logic rather than statistical co-expression remains unclear. Sparse autoencoders (SAEs) can resolve superposition in neural networks by decomposing dense activations into interpretable features, yet they have not been systematically appli...

Topics

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

Identifiers and source

Literature Corpus work: ca90eba0-1037-5a6e-b07d-0bd2b2cd8326
DOI: 10.21203/rs.3.rs-9082479/v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Sparse autoencoders reveal organized biological knowledge but minimal regulatory logic in single-cell foundation models: a comparative atlas of Geneformer and scGPTDOI 10.21203/rs.3.rs-9082479/v1Read

Regulatory Logic in Larger Single-Cell Foundation Models: A Sparse Autoencoder Comparison of scFoundation and GeneformerSemantic proximity Read

RegFormer: A Single-Cell Foundation Model Powered by Gene Regulatory HierarchiesSemantic proximity Read

AENetMoX: Fast and Precise scRNA-seq Gene Regulatory Network Inference Through Multi-Modal Feature Integration in Brain OrganoidsSemantic proximity Read

Learning the spatial cell-cell communication network to decode multi-channel signaling and predict network-hub vulnerabilities with MOSANICSemantic proximity Read

Multimodal foundation model predicts zero-shot functional perturbations and cell fate dynamicsSemantic proximity Read

scEMB: Learning context representation of genes based on large-scale single-cell transcriptomicsSemantic proximity Read

Decoding Conserved and Tissue-Specific Cellular Stress Programs from Single-Cell Foundation Models Using Sparse AutoencodersSemantic proximity Read

OmniCell: Unified Foundation Modelling of Single-Cell and Spatial Transcriptomics for Cellular and Molecular InsightsSemantic proximity Read

Large Scale Foundation Model on Single-cell TranscriptomicsSemantic proximity Read

Inhibitory Knowledge Flow from E-Communities of Practice Stabilizes Causal Gene Circuit Architectures in Single-Cell Foundation ModelsSemantic proximity Read

BoYueGRN: Zero-shot causal discovery of directed gene regulatory networks from single-cell transcriptomes via amortized inference over synthetic structural causal modelsSemantic proximity Read