Genome-wide association studies (GWAS) have been successful in identifying over 100 loci associated with Parkinson’s disease (PD) susceptibility. However, the majority of these studies have focused on European cohorts with few including diverse ancestries. Using genotyped and imputed data from 691 South African PD cases and 826 controls, we conducted a conventional GWAS, two local ancestry GWAS (LA-GWAS) approaches (one using local ancestry as a covariate and the other separating the dosage per ancestry), and an association analysis to identify regions of homozygosity associated with PD status. Furthermore, we replicated these findings using another admixed population, a Latin American cohort (LARGE-PD). The ancestry inference suggested that the South African cohort is admixed from five populations, including African (AFR), European (EUR), Malaysian (MAL), Nama (NAMA), and South Asian (SAS), though with varying accuracy levels. The conventional GWAS successfully identified one locus (rs17098735-T) with genome-wide significance (p-value: 1.23×10-8; beta= 2.286; SE= 0.401). Within the local ancestry window of the top GWAS hit, among individuals carrying the variant, 86.7% had AFR ancestry, 11% NAMA ancestry, and 2.2% MAL ancestry, with no EUR or SAS ancestry observed, highlighting a potential ancestry-specific genetic risk factor. Three lead loci were replicated in the LARGE-PD cohort. LA-GWAS using the Cochran-Armitage trend test identified 35 lead SNPs above suggestive significance after multiple test correction. Tractor-based approaches identified three lead loci when analyzing all five ancestry components jointly, as well as three additional lead loci in the AFR-only component, highlighting ancestry-specific loci that may contribute to genetic risk in diverse populations. For the LA-GWAS, one independent locus was replicated in LARGE-PD. Our findings suggest ancestry specificity in PD risk and underscores the importance of including diverse populations in genetics research. The study contributes towards a global understanding of the genetic etiology underlying PD.

The LRRK2 p.Gly2019Ser pathogenic variant has reduced penetrance and presents a wide range of age at onset (AAO) in patients with Parkinson’s disease (PD). We aim to elucidate differences in the cumulative incidence of LRRK2 p.Gly2019Ser-related PD (LRRK2-PD) between ancestries and countries. We included N=922 unrelated LRRK2 p.Gly2019Ser variant carriers (affected: N=762, unaffected: N=160) from the Global Parkinson’s Genetics Program (GP2) in addition to cohorts recruited from the Israeli Ashkenazi Jewish and Tunisian Arab-Berber population. The p.Gly2019Ser variant was present in five ancestries: Ashkenazi Jewish (N=534), North African (N=223), European (N=132), Middle Eastern (N=19) and Latino and Indigenous people of the Americas (N=14). In addition to ancestry derived from the genetic data, we assessed the country of origin in our analysis. The Cox proportional-hazards model and Kaplan-Meier analysis were applied to examine differences in cumulative incidence. All analyses were adjusted for biological sex, and the outcome variable was AAO, including affected and unaffected variant carriers with right censoring for affection status, and all analysis were exploratory. The median AAO of LRRK2-PD was five years younger in the North African (HR=1.48, 95% CI: 1.18-1.86, p=7.0×10−4) compared to the European ancestry group. In contrast, the median AAO was five years older in the Ashkenazi Jewish (HR=0.61, 95% CI: 0.50-0.75, p=4.0×10−6) compared to the European ancestry group. Additionally, patients from Israel (HR=1.59, 95% CI: 1.30-1.39, p=4.0×10−6) and Tunisia (HR=2.57, 95% CI: 2.16-3.06, p<2.0×10−16) had a median 5-year and 10-year younger AAO compared to patients from the USA, respectively. Lastly, when focusing only on individuals with an Ashkenazi Jewish background, patients from Israel still had a younger AAO than those from the USA (HR=1.82, 95% CI: 1.48-2.24, p=1.5×10−8). Analogously, assessing only patients from the USA, the Ashkenazi Jewish ancestry group still had an older AAO than the European ancestry group (HR=0.51, 95% CI: 0.39-0.67, p=1.3×10−6).

The genomic landscape of the Indian population, particularly for age-related disorders like Parkinson’s disease (PD) remains underrepresented in global research. Genetic variability in PD has been studied predominantly in European populations, offering limited insights into its role within the Indian population. To address this gap, we conducted the first pan-India genomic survey of PD involving 4,806 cases and 6,364 controls, complemented by a meta-analysis integrating summary statistics from a multi-ancestry PD meta-analysis (N=611,485). We further leveraged RNA-sequencing data from lymphoblastoid cell lines of 731 individuals from the 1000 Genomes project to evaluate the expression of key loci across global populations. Our findings reveal a higher genetic burden of PD in the Indian population compared to Europeans, accounting for ∼30% of the previously unexplained heritability. Thirteen genome-wide significant loci were identified, including two novel loci, with an additional three loci uncovered through meta-analysis. Polygenic risk score analysis showed moderate transferability from European populations. Our results highlight the importance of genetic loci in immune function, lipid metabolism and SNCA aggregation in PD pathogenesis, with gene expression variability emphasizing population-specific differences. We also established South Asia’s largest PD biobank, providing a foundation for patient-centric approaches to PD research and treatment in India.

Mitochondrial dysfunction is a key player in Parkinson’s disease (PD) pathogenesis. Mitochondrial polygenic scores (MGS) may be associated with PD but require validation across diverse populations. To validate the association between the MGS, PD status and age-at-onset (AAO) in idiopathic and LRRK2-PD across various ancestries. We analyzed data from 17,129 PD patients and 13,872 healthy individuals across 10 ancestries within the Global Parkinson’s Disease Genetic Program. We used regression models to assess the association between MGS, PD status and AAO. The MGS was associated with iPD in Europeans (β=0.19, SE=0.02, p<2.0×10−16) and Ashkenazi Jews (β=0.26, p=3.7×10-4) but not in other populations. Additionally, the MGS was strongly associated with LRRK2-PD status (β=0.82, p=2.0×10−16). No associations with AAO were observed. The MGS is robustly associated with iPD status in Europeans and Ashkenazi Jews and with LRRK2-PD status. Population-specific MGS are needed to improve accuracy in other ancestries.

We conducted a meta-analysis of Parkinson’s disease genome-wide association study summary statistics, stratified by source (clinically-recruited case-control cohorts versus population biobanks) and by general European versus European isolate ancestries. This study included 63,555 cases, 17,700 proxy cases with a family history of Parkinson’s disease, and 1,746,386 controls, making it the largest investigation of Parkinson’s disease genetic risk to date. The final combined cross-European meta-analysis identified 134 risk loci (59 novel), with a total of 157 independent signals, significantly expanding our understanding of Parkinson’s disease risk. Multi-omic data integration revealed that expression of the nominated risk genes are highly enriched in brain tissues, particularly in neuronal and astrocyte cell types. Additionally, we prioritized 33 high-confidence genes across these 134 loci for future follow-up studies.

The catechol-O-methyltransferase (COMT) gene is involved in brain catecholamine metabolism, but its association with Parkinson’s disease (PD) risk remains unclear. To investigate the relationship between COMT genetic variants and PD risk across diverse ancestries. We analyzed COMT variants in 2,251 PD patients and 2,835 controls of European descent using whole-genome sequencing from the Accelerating Medicines Partnership-Parkinson Disease (AMP-PD), along with 20,427 PD patients and 11,837 controls from 10 ancestries using genotyping data from the Global Parkinson’s Genetics Program (GP2). Utilizing the largest case-control datasets to date, no significant enrichment of COMT risk alleles in PD patients was observed across any ancestry group after correcting for multiple testing. Among Europeans, no correlations with cognitive decline, motor function, motor complications, or time to LID onset were observed. These findings emphasize the need for larger, diverse cohorts to confirm the role of COMT in PD development and progression.

This study evaluated genotyping success of the NeuroBooster array (NBA) and determined the frequencies of pathogenic variants across ancestries. We analyzed the presence and allele frequency of 34 pathogenic variants in 28,710 PD cases, 9,614 other neurodegenerative disorder cases, and 15,821 controls across 11 ancestries within the Global Parkinson’s Genetics Program dataset. Of these, 25 were genotyped on NBA and cluster plots were used to assess their quality. Genes previously predicted to have high or very high confidence of causing PD tend to have more pathogenic variants and are present across ancestry groups. Twenty-five of the 34 pathogenic variants were typed by the NBA array and classified “good” (n=12), “medium” (n=4), and “bad” (n=9) variants. Our results confirm the likelihood that established PD genes are pathogenic and highlight the importance of ancestrally diverse research in PD. We also show the usefulness of the NBA as a reliable tool for genotyping of rare variants for PD.

Copy Number Variations (CNVs) play pivotal roles in the etiology of complex diseases and are variable across diverse populations. Understanding the association between CNVs and disease susceptibility is significant in disease genetics research and often requires analysis of large sample sizes. One of the most cost-effective and scalable methods for detecting CNVs is based on normalized signal intensity values, such as Log R Ratio (LRR) and B Allele Frequency (BAF), from Illumina genotyping arrays. In this study, we present CNV-Finder, a novel pipeline integrating deep learning techniques on array data, specifically a Long Short-Term Memory (LSTM) network, to expedite the large-scale identification of CNVs within predefined genomic regions. This facilitates efficient prioritization of samples for time-consuming or costly subsequent analyses such as Multiplex Ligation-dependent Probe Amplification (MLPA), short-read, and long-read whole genome sequencing. We incorporate four genes to establish our methods—Parkin (PRKN), Leucine Rich Repeat And Ig Domain Containing 2 (LINGO2), Microtubule Associated Protein Tau (MAPT), and alpha-Synuclein (SNCA)—which may be relevant to neurological diseases such as Alzheimer’s disease (AD), Parkinson’s disease (PD), Progressive Supranuclear Palsy (PSP), or related disorders such as essential tremor (ET). By training our models on expert-annotated samples and validating them across diverse cohorts, including those from the Global Parkinson’s Genetics Program (GP2) and additional dementia-specific databases, we demonstrate the efficacy of CNV-Finder in accurately detecting deletions and duplications. Our pipeline outputs app-compatible files for visualization within CNV-Finder’s interactive web application. This interface enables researchers to review predictions and filter displayed samples by model prediction values, LRR range, and variant count in order to explore or confirm results. Our pipeline integrates this human feedback to enhance model performance and reduce false positive rates. Through a series of comprehensive analyses and validations using visual inspection, MLPA, short-read, and long-read sequencing data, we demonstrate the robustness and adaptability of CNV-Finder in identifying CNVs with regions of varied size, probe density, and noise. Our findings highlight the significance of contextual understanding and human expertise in enhancing the precision of CNV identification, particularly in complex genomic regions like 17q21.31. The CNV-Finder pipeline is a scalable, publicly available resource for the scientific community, available on GitHub (https://github.com/GP2code/CNV-Finder; DOI 10.5281/zenodo.14182563). CNV-Finder not only expedites accurate candidate identification but also significantly reduces the manual workload for researchers, enabling future targeted validation and downstream analyses in regions or phenotypes of interest.

We conducted the first large-scale, multi-ancestral investigation of Parkinson’s disease (PD) to examine the impact of genome-wide homozygosity on disease risk and age at onset. Using genotyping, imputed, and whole-genome sequencing (WGS) data from 16,599 PD cases and 13,585 controls across nine ancestral populations from the Global Parkinson’s Genetics Program, we aimed to identify novel regions of homozygosity contributing to PD heritability. Our findings suggest that ROH regions contribute to PD heritability in a global context, with a portion attributed to recessive allelic architecture. We developed an open-science framework for unbiased homozygosity mapping. Future studies should use larger, diverse cohorts and WGS data to uncover rare recessive variants linked to PD susceptibility.

Emerging evidence suggests that the genetic architecture of Alzheimer’s (AD) and Parkinson’s diseases (PD) risk varies across ancestries. This study seeks to explore distinct and universal genetic targets across individuals of Latino, African/African Admixed, East Asian, and European populations by implementing Population Attributable Risk (PAR) comparisons on summary statistics from genome-wide association studies (GWAS). PAR was calculated for the most significant disease variants using summary statistics derived from select multi-ancestry GWAS meta-analyses, followed by fine-mapping analysis to validate genetic contribution of disease variants to European, African/African Admixed, East Asian, and Latino individuals. For both AD, APOE4 PAR estimates were universally high across all ancestries, with TSPAN14 and PICALM emerging as other common targets. Attributable risk varied across PD-related major risk loci including variation nearby GBA1 and LRRK2. In contrast, SNCA, MCCC1, VPS13C, and MAPT loci demonstrated comparable attributable risk across ancestries. This cross-ancestry evaluation of PAR reinforces the genetic heterogeneity of AD and PD. In consideration of the complex etiology of these diseases, these findings may inform the strategic prioritization of therapeutic targets and improve global health outcomes.