4404 papers
This paper systematically investigates the manifold hypothesis in protein language models (pLMs) using a comprehensive battery of geometric and topological diagnostics, concluding that unlike single-cell foundation models, pLM representations are predominantly high-dimensional and linear rather than organized around low-dimensional manifolds.
This empirical study reveals that while biological large language models (SCFMs) do encode a real, low-dimensional, and partially linearized geometric structure for biological knowledge, this structure is modest, often nonlinearly entangled, and largely overlaps with what classical expression-based methods like PCA can already recover, falling short of the crisp, regular geometry observed in natural language models.
This paper presents a unified benchmark of eleven virtual cell models, revealing that while deep learning approaches significantly outperform baselines in representation tasks and combinatorial perturbation prediction, they generally fail to surpass simple linear models in predicting unseen single-gene perturbations, suggesting the field has achieved a "virtual microscope" for cell state analysis rather than a true "virtual simulator" for causal perturbation dynamics.
This study demonstrates that while sparse autoencoders effectively extract monosemantic, biologically interpretable features like sequence motifs and reading frames from DNA foundation models, they reveal a significant scarcity of features encoding complex regulatory logic, suggesting current models capture a genomic dictionary rather than a grammar engine.
This paper presents a unified empirical analysis across six biological data modalities demonstrating that finite-sample bias severely distorts information-theoretic estimates, and argues that reporting entropy and mutual information without specifying the estimator and sample size is indefensible given that bias-corrected methods reveal distinct biological signals and correct systematic underestimation of entropy and overestimation of information content.
This empirical study demonstrates that for biological-age prediction across diverse ageing cohorts, pre-transformer models (such as penalized regression and gradient boosting) generally outperform or match attention-based transformers in accuracy, data efficiency, cross-cohort transfer, and mortality risk stratification, establishing them as the rational default for longevity science while positioning transformers as a specialized tool only viable with exceptionally large datasets.
This empirical study demonstrates that for blood-based aging clocks, the choice of prediction target (mortality-derived phenotypic age versus chronological age) overwhelmingly determines biomarker validity and mortality association, rendering increases in deep learning model capacity largely ineffective for improving these critical outcomes.
This study proposes that Alzheimer's disease results from the aberrant epigenetic reactivation of silenced embryonic actin-myosin oscillatory pathways in aging neurons and glia, creating mechanical instability that drives axonal "dying-back" and synaptic retraction rather than simple passive proteotoxicity.
This paper has been withdrawn by its authors to comply with an institutional preprint policy, and they explicitly request that the work not be cited.
This study identifies REEP2 as a critical mediator of EMT-driven lung cancer progression, demonstrating that the ZEB1/miR-183/193a axis upregulates REEP2 to enhance ER-to-Golgi trafficking and the secretion of pro-tumorigenic factors, thereby promoting tumor metastasis and immunosuppression.