Abstract
Background Genome-wide association studies (GWAS) have identified numerous risk loci for Parkinson’s disease, yet identifying causal genes and mechanisms remains challenging due to non-coding associations and complex linkage disequilibrium.
Methods We prioritized genes within 147 GWAS loci using an XGBoost machine-learning model trained on 285 multi-omic features, including brain-specific eQTLs and single-cell expression. Prioritized candidates underwent gene- and domain-level rare variant burden analysis via optimal sequenced kernel association test (SKAT-O) across the Accelerating Medicines Partnership for Parkinson’s Disease and UK Biobank cohorts (N case = 6,435, N proxy = 13,889, N control = 343,160).
Findings The model prioritized 406 genes with 48 features contributing to the prioritization. Meta-analysis of rare variants at the gene and domain levels replicated established associations (GBA1, LRRK2) and identified six novel potential risk genes ANKRD27 driven by p.Arg21Cys, UBXN2A, FAM171A1, ERCC8, BNC2, and ADNP. Domain-level analysis uniquely uncovered a significant association in the zinc-finger domain of ADNP, which was masked in gene-level tests. Additionally, at cohort-level, one novel gene was nominated: LRRC45 driven by p.Met607ArgfsTer57.
Interpretation Integrating machine-learning prioritization with gene- and domain-level burden testing identified novel genes potentially involved in Parkinson’s disease, with further validation needed to elucidate causality and mechanisms.
Funding This study was supported by grants from the Galen and Hilary Weston Foundation, the Michael J. Fox Foundation, the Canadian Consortium on Neurodegeneration in Aging (CCNA), the Canada First Research Excellence Fund (CFREF). Lastly, the study received contributions from the G-Can (GBA1-Canada) Initiative. G-Can is supported by The Hilary and Galen Weston Foundation, Silverstein Foundation, and J. Sebastian van Berkom and Ghislaine Saucier. S.C.P. is supported by a Canadian Institutes of Health Research (CIHR) Canada Graduate Scholarship – Doctoral (CGS-D) award. Z.G.-O. holds a Fonds de recherche du Québec - Santé (FRQS) Chercheurs-boursiers award and is a William Dawson Scholar.
Competing Interest Statement
Z. G.-O. received consultancy fees from Lysosomal Therapeutics Inc. (LTI), Idorsia, Prevail Therapeutics, Inceptions Sciences (now Ventus), Neuron23, Handl Therapeutics, UCB, Capsida, Vanqua Bio, Congruence Therapeutics, Ono Therapeutics, Denali, Bial Biotech, Bial, EG427, Takeda, Jazz Pharmaceuticals, Simcere, Guidepoint, Lighthouse, and Deerfield. K.S. received consultancy fees from Acurex.
Funding Statement
This work was supported by grants from the Galen and Hilary Weston Foundation, the Michael J. Fox Foundation, the Canadian Consortium on Neurodegeneration in Aging (CCNA), and the Canada First Research Excellence Fund (CFREF), awarded to McGill University for the Healthy Brains for Healthy Lives (HBHL) initiative. We also acknowledge contributions from the G-Can (GBA1-Canada) Initiative, an open-science collaboration dedicated to GBA1-associated neurodegeneration. G-Can is supported by the Galen and Hilary Weston Foundation, the Silverstein Foundation, and J. Sebastian van Berkom and Ghislaine Saucier. S.C.P. is supported by a Canadian Institutes of Health Research (CIHR) Canada Graduate Scholarship - Doctoral (CGS-D) award. Z.G.-O. holds a Fonds de recherche du Québec - Santé (FRQS) Chercheurs-boursiers award and is a William Dawson Scholar.
Author Declarations
I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained.
Yes
The details of the IRB/oversight body that provided approval or exemption for the research described are given below:
Institutional Review Board of McGill University gave ethical approval for this work (IRB Study Number 21-11-023).
I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals.
Yes
I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance).
Yes
I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable.
Yes
Data Availability
The WGS cohort accessed and analysed in this study for the purposes of the burden analysis are accessible (conditional to institutional approval) at: AMP-PD Knowledge Platform (https://www.amp-pd.org) and the UK Biobank Research Analysis Platform. All information directly relevant to the study are included in the text and the supplementary files of this article, except for the full range of analysed variants, which have been made accessible online at the link: https://doi.org/10.5281/zenodo.18285356. All the scripts used in the machine learning prioritization and burden analyses are shared at the GitHub repository link: https://github.com/gan-orlab/ML_and_Burden_PD.
Abbreviations
- AMP-PD
- Accelerating Medicines Partnership for Parkinson’s Disease
- eQTL
- Expression quantitative trait loci
- GWAS
- Genome-wide association studies
- ML
- Machine learning
- PD
- Parkinson’s disease
- SKAT-O
- Optimal sequenced kernel association test
- UKBB
- UK Biobank





