OmicFormer, a Transformer-based architecture that embeds two complementary statistical priors, i.e., feature-label associations and feature-feature dependencies, directly into its representation learning, provides an interpretable and generalizable foundation for omics-based precision medicine.
Genetic prediction of complex phenotypes typically relies on additive linear models, which scale well but cannot capture non-additive effects or deeply integrate molecular and clinical data. Domain-specific neural networks have driven advances in images, text, and other modalities, but genome-scale neural networks rema...
J. Upmeier zu Belzen, L. Arnoldt, N. Hollmann et al.· medRxiv· 0 citations
ProMetNet is presented, a biochemically constrained framework that incorporates pathway‐derived connectivity from the Reactome database into neural network architecture that consistently outperforms evaluated multi‐omics integration methods, maintaining high discriminative performance under substantial data downsamplin...
Ming-Hui Zhao, Na Zhou, Ruo-Tong Liu et al.· Advancement of science· 0 citations
It is shown that human genetics provides an orthogonal layer to surface implicitly learned biological concepts and otherwise hidden risk factors to train the next generation of EHR-based foundation models that likely benefit most from the addition of targeted biomarkers to advance personalised medicine.
M. Pietzner, W. Zeng, L. Kohleick et al.· medRxiv· 0 citations
MB-SupCon-cont improves prediction accuracy by incorporating a generalized contrastive loss function that defines similarity and dissimilarity for continuous responses using three distance-based weighting methods, and provides superior representation learning and improves data visualization in lower-dimensional spaces.
Sen Yang, Shidan Wang, Yiqing Wang et al.· Frontiers in microbiomes· 0 citations
NITROGEN is proposed, an imputation-free transformer that jointly models within-patient feature dependencies and between-patient relational structure through masked and intersample attention, enabling robust multimodal learning directly from partially observed records, and shows robust probability calibration and uncer...
Christelle Schneuwly Diaz, Narmina Baghirova, Thanh Duy Vu et al.· arXiv.org· 0 citations
Machine learning is a ubiquitous part of the modern neuroimaging toolkit, particularly for research aimed towards precision medicine goals of improving individual-level diagnosis and treatment. However, the high dimensionality of neuroimaging data poses significant challenges for constructing interpretable predictive m...
