Discovery of a Hematopoietic Manifold in scGPT Yields a Method for Extracting Performant Algorithms from Biological Foundation Model Internals
We report the discovery and extraction of a compact hematopoietic algorithm from the single-cell foundation model scGPT, to our knowledge the first biologically useful, competitive algorithm extracted from a foundation model via mechanistic interpretability. We show that scGPT internally encodes a compact hematopoietic manifold with significant developmental branch structure, validated on a strict non-overlap Tabula Sapiens external panel and confirmed via frozen-head zero-shot transfer to an independent multi-donor immune panel. To isolate this geometry, we introduce a general three-stage extraction method consisting of direct operator export from frozen attention weights, a lightweight learned adaptor, and a task-specific readout, producing a standalone algorithm without target-dataset retraining. In 88-split donor-holdout benchmarks against scVI, Palantir, DPT, CellTypist, PCA, and raw-expression baselines, the extracted algorithm achieves the strongest pseudotime-depth ordering and leads on key subtype endpoints (CD4/CD8 AUROC 0.867, mono/macro AUROC 0.951). Compared to standard probing of frozen scGPT embeddings with a 3-layer MLP, the extracted head is BH-significantly better on 6/8 classification endpoints while completing a full 12-split evaluation campaign 34.5x faster with approximately 1000x fewer trainable parameters. The exported operator compresses from three pooled attention heads to a single head without statistically significant loss, and further to a rank-64 surrogate. Mechanistic interpretability of the compact operator reveals a concentrated four-factor core explaining 66.2% of ablation impact, with factors resolving into explicit T/lymphoid, B/plasma, granulocytic, and monocyte/macrophage gene programs. A supplementary second-manifold validation (intercellular communication geometry) confirms that the extraction method generalizes beyond hematopoiesis.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
What Topological and Geometric Structure Do Biological Foundation Models Learn? Evidence from 141 Hypotheses
When biological foundation models such as scGPT and Geneformer process single-cell gene expression, what geometric and topological structure forms in their internal representations? Is that structure biologically meaning…
Sparse autoencoders reveal organized biological knowledge but minimal regulatory logic in single-cell foundation models: a comparative atlas of Geneformer and scGPT
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 a…
Causal Circuit Tracing Reveals Distinct Computational Architectures in Single-Cell Foundation Models: Inhibitory Dominance, Biological Coherence, and Cross-Model Convergence
Motivation: Sparse autoencoders (SAEs) decompose foundation model activations into interpretable features, but causal feature-to-feature interactions across network depth remain unknown for biological foundation models. …
Atlas-based Manifold Representations for Interpretable Riemannian Machine Learning
Despite the popularity of the manifold hypothesis, current manifold-learning methods do not support machine learning directly on the latent $d$-dimensional data manifold, as they primarily aim to perform dimensionality r…
Dimensionality ReductionIntegrating Single-Cell Foundation Models with Graph Neural Networks for Drug Response Prediction
AI-driven drug response prediction holds great promise for advancing personalized cancer treatment. However, the inherent heterogenity of cancer and high cost of data generation make accurate prediction challenging. In t…
Drug Response PredictionPrediction