4244 papers
Despite significant architectural and parameter differences, single-cell foundation models share a small, robust, and biologically interpretable low-dimensional core that outperforms their individual full embeddings in functional retrieval tasks, revealing that most of their representational capacity is private and biologically inert.
This paper demonstrates that the eligibility trace is the singular, non-negotiable temporal mechanism required for emergent differentiation in reward-modulated spiking networks, as its removal abolishes order while its extension directly governs the degree of differentiation regardless of network scale.
This paper reveals that the leading spectral axes of single-cell foundation models are predominantly confounded by gene expression abundance rather than biological meaning, demonstrating that removing these axes improves retrieval performance and establishing a model-dependent compaction law that guides optimal dimensionality reduction for biological tasks.
This study demonstrates that while complex machine learning models can more accurately predict chronological age from blood chemistry, their resulting "age acceleration" metrics are less effective at predicting mortality than simpler models or biomarkers directly trained on death outcomes, revealing that the training target rather than the model complexity or biomarkers is the critical factor in developing useful biological age predictors.
This study demonstrates that in single-cell donor-level analyses, controlling for cell-type composition often removes negligible signal because expression profiles retain most trait information even when composition is fixed, and that the perceived strength of composition as a confounder is highly dependent on the resolution of the cell-type annotation used.
This paper presents Spatial GNN Analyzer, an AI-powered framework that integrates Graph Neural Networks with spatial transcriptomics data to enable robust, biologically coherent spatial domain identification and interactive visualization, demonstrating superior performance in preserving tissue architecture compared to conventional non-spatial methods.
This study reveals that grazing alters *Stipa grandis* root exudate profiles to drive asymmetric microbial responses, where short-term grazing promotes bacterial growth via energy-rich lipids while long-term grazing sustains fungal associations through defensive metabolites, with metabolite network positions constrained by thermodynamic properties in bacteria but by molecular structural complexity in fungi.
This study reveals that evolutionary conservation frequently generates false-positive predictions of deleteriousness for missense mutations, a failure mode that is strongly and independently predicted by high solvent accessibility and conformational flexibility across ten deep mutational scanning datasets.
This study demonstrates that targeting cancer-intrinsic protein neddylation overcomes acquired resistance to immunotherapy in JAK/STAT-deficient tumors by stabilizing chromatin-bound cGAS, thereby restoring inflammatory signaling and enhancing immune cell infiltration to sensitize tumors to PD1 blockade.
This study introduces the MiXeR-SV tool to demonstrate that structural variants, despite comprising a small fraction of genomic variants, significantly contribute to the heritability of numerous complex traits—particularly in hematological, metabolic, and cancer phenotypes—thereby helping to resolve the gap between SNP-based and twin heritability estimates.