Machine learning in Alzheimer’s disease genetics
Traditional statistical approaches have advanced our understanding of the genetics of complex diseases, yet are limited to linear additive models. Here we applied machine learning (ML) to genome-wide data from 41,686 individuals in the largest European consortium on Alzheimer’s disease (AD) to inves...
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Article (Journal) |
| Language: | English |
| Published: |
22 July 2025
|
| In: |
Nature Communications
Year: 2025, Volume: 16, Pages: 1-16 |
| ISSN: | 2041-1723 |
| DOI: | 10.1038/s41467-025-61650-z |
| Online Access: | Verlag, kostenfrei: https://pmc.ncbi.nlm.nih.gov/articles/PMC12280214/ Verlag, kostenfrei: https://www.nature.com/articles/s41467-025-61650-z Verlag, kostenfrei, Volltext: https://doi.org/10.1038/s41467-025-61650-z |
| Author Notes: | Matthew Bracher-Smith, Federico Melograna, Brittany Ulm, Céline Bellenguez, Benjamin Grenier-Boley, Diane Duroux, Alejo J. Nevado, Peter Holmans, Betty M. Tijms, Marc Hulsman, Itziar de Rojas, Rafael Campos-Martin, Sven van der Lee, Atahualpa Castillo, Fahri Küçükali, Oliver Peters, Anja Schneider, Martin Dichgans, Dan Rujescu, Norbert Scherbaum, Jürgen Deckert, Steffi Riedel-Heller, Lucrezia Hausner, Laura Molina-Porcel, Emrah Düzel, Timo Grimmer, Jens Wiltfang, Stefanie Heilmann-Heimbach, Susanne Moebus, Thomas Tegos, Nikolaos Scarmeas, Oriol Dols-Icardo, Fermin Moreno, Jordi Pérez-Tur, María J. Bullido, Pau Pastor, Raquel Sánchez-Valle, Victoria Álvarez, Mercè Boada, Pablo García-González, Raquel Puerta, Pablo Mir, Luis M. Real, Gerard Piñol-Ripoll, Jose María García-Alberca, Eloy Rodriguez-Rodriguez, Hilkka Soininen, Sami Heikkinen, Alexandre de Mendonça, Shima Mehrabian, Latchezar Traykov, Jakub Hort, Martin Vyhnalek, Nicolai Sandau, Jesper Qvist Thomassen, Yolande A.L. Pijnenburg, Henne Holstege, John van Swieten, Inez Ramakers, Frans Verhey, Philip Scheltens, Caroline Graff, Goran Papenberg, Vilmantas Giedraitis, Julie Williams, Philippe Amouyel, Anne Boland, Jean-François Deleuze, Gael Nicolas, Carole Dufouil, Florence Pasquier, Olivier Hanon, Stéphanie Debette, Edna Grünblatt, Julius Popp, Roberta Ghidoni, Daniela Galimberti, Beatrice Arosio, Patrizia Mecocci, Vincenzo Solfrizzi, Lucilla Parnetti, Alessio Squassina, Lucio Tremolizzo, Barbara Borroni, Michael Wagner, Benedetta Nacmias, Marco Spallazzi, Davide Seripa, Innocenzo Rainero, Antonio Daniele, Fabrizio Piras, Carlo Masullo, Giacomina Rossi, Frank Jessen, Patrick Kehoe, Tsolaki Magda, Pascual Sánchez-Juan, Kristel Sleegers, Martin Ingelsson, Mikko Hiltunen, Rebecca Sims, Wiesje van der Flier, Ole A. Andreassen, Agustín Ruiz, Alfredo Ramirez, Ruth Frikke-Schmidt, Najaf Amin, Gennady Roshchupkin, Jean-Charles Lambert, Kristel Van Steen, Cornelia van Duijn & Valentina Escott-Price |
| Summary: | Traditional statistical approaches have advanced our understanding of the genetics of complex diseases, yet are limited to linear additive models. Here we applied machine learning (ML) to genome-wide data from 41,686 individuals in the largest European consortium on Alzheimer’s disease (AD) to investigate the effectiveness of various ML algorithms in replicating known findings, discovering novel loci, and predicting individuals at risk. We utilised Gradient Boosting Machines (GBMs), biological pathway-informed Neural Networks (NNs), and Model-based Multifactor Dimensionality Reduction (MB-MDR) models. ML approaches successfully captured all genome-wide significant genetic variants identified in the training set and 22% of associations from larger meta-analyses. They highlight 6 novel loci which replicate in an external dataset, including variants which map to ARHGAP25, LY6H, COG7, SOD1 and ZNF597. They further identify novel association in AP4E1, refining the genetic landscape of the known SPPL2A locus. Our results demonstrate that machine learning methods can achieve predictive performance comparable to classical approaches in genetic epidemiology and have the potential to uncover novel loci that remain undetected by traditional GWAS. These insights provide a complementary avenue for advancing the understanding of AD genetics. |
|---|---|
| Item Description: | Gesehen am 02.12.2025 |
| Physical Description: | Online Resource |
| ISSN: | 2041-1723 |
| DOI: | 10.1038/s41467-025-61650-z |