End-to-end deep learning methods for genetic risk prediction of schizophrenia
Nora Verplaetse, Yves Moreau, Daniele Raimondi · 2026
Read End-to-end deep learning methods for genetic risk prediction of schizophrenia by Nora Verplaetse, Yves Moreau, Daniele Raimondi (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13631334.txt.
Description
Nora Verplaetse, Yves Moreau, and Daniele Raimondi wrote this 2026 paper on deep learning for predicting schizophrenia risk from genes. Schizophrenia runs strongly in families but rests on a tangled genetic base. Past genome scans found many risk spots, yet clinic models mostly assume each gene adds a simple, straight-line effect.
That view misses tangled effects between genes, called epistasis. The team applied end-to-end neural networks to whole exome data, the protein-coding part of the genome. Their cohort held 6,135 cases and 6,245 controls. With enough data, the nonlinear networks beat the standard additive models by a clear margin. The authors add explainable AI tools and biological priors to see how the models decide. Their case is that gene interactions belong in any serious risk model.
Catalog record
| CREATOR | Nora Verplaetse, Yves Moreau, Daniele Raimondi |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 91 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |