Topological stratification of continuous genetic variation in large biobanks

Diaz-Papkovich, Alex; Zabad, Shadi; Ben-Eghan, Chief; Anderson-Trocmé, Luke; Femerling, Georgette; Nathan, Vikram; Patel, Jenisha; Gravel, Simon

doi:10.1101/2023.07.06.548007

Cited by 3 publications

(1 citation statement)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In future work, a more sophisticated approach based on genetic clustering methods (e.g. [Prive, 2022, Diaz-Papkovich et al, 2023]) could yield further improvements.…”

Section: Experimentationmentioning

confidence: 99%

Multi-ancestry polygenic risk scores using phylogenetic regularization

Layne,

Zabad,

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

Accurately predicting phenotype using genotype across diverse ancestry groups remains a significant challenge in human genetics. Many state-of-the-art polygenic risk score models are known to have difficulty generalizing to genetic ancestries that are not well represented in their training set. To address this issue, we present a novel machine learning method for fitting genetic effect sizes across multiple ancestry groups simultaneously, while leveraging prior knowledge of the evolutionary relationships among them. We introduce DendroPRS, a machine learning model where SNP effect sizes are allowed to evolve along the branches of the phylogenetic tree capturing the relationship among populations. DendroPRS outperforms existing approaches at two important genotype-to-phenotype prediction tasks: expression QTL analysis and polygenic risk scores. We also demonstrate that our method can be useful for multi-ancestry modelling, both by fitting population-specific effect sizes and by more accurately accounting for covariate effects across groups. We additionally find a subset of genes where there is strong evidence that an ancestry-specific approach improves eQTL modelling.

show abstract

“…In future work, a more sophisticated approach based on genetic clustering methods (e.g. [Prive, 2022, Diaz-Papkovich et al, 2023]) could yield further improvements.…”

Section: Experimentationmentioning

confidence: 99%

Multi-ancestry polygenic risk scores using phylogenetic regularization

Layne,

Zabad,

et al. 2024

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Rare diseases load through the study of a regional population

Michel,

Moreau,

Gagnon

et al. 2024

Preprint

View full text Add to dashboard Cite

Rare genetic diseases impact many people worldwide and are challenging to diagnose. In this study, we introduce a novel regional population cohort approach to identify pathogenic variants that occur more frequently within specific populations and are of clinical interest. We utilized a cohort from Quebec, including the Saguenay-Lac-Saint-Jean region, which is known for its founder effect and higher frequency of certain pathogenic variants. By analyzing both the frequency of these variants and their origin through shared identical-by-descent segments, we validated 38 variants previously reported as being more common due to the founder effect. Additionally, we identified 42 unreported founder variants in Quebec or the Saguenay-Lac-Saint-Jean, some with carrier rates estimates as high as 1/22. We also observed a greater deleterious mutational load for the studied variants in individuals from the Saguenay-Lac-Saint-Jean compared to other urban Quebec regions. These findings were brought to the clinic where 12 pathogenic variants were detected in patients, including 3 that are responsible for very severe diseases and could be considered for inclusion in a carrier test for the Saguenay-Lac-Saint-Jean population. This study highlights the potential underestimation of rare disease prevalence and presents a population-based approach that could aid clinicians in their diagnostic efforts and patients' management.

show abstract

The multi-scale complexity of human genetic variation beyond continental groups

Palma-Martínez,

Posadas-García,

López-Ángeles

et al. 2024

Preprint

View full text Add to dashboard Cite

Traditional clustering and visualization approaches in human genetics often operate under frameworks that assume inherent, discrete groupings1,2. These methods can inadvertently simplify multifaceted relationships, functioning to entrench the idea of typological groups3. We introduce a network-based pipeline and visualization tool grounded in relational thinking4, which constructs networks from a variety of genetic similarity metrics. We identify communities at multiple resolutions, departing from typological models of analysis and interpretation that categorize individuals into a (predefined) number of sets. We applied our pipeline to a dataset merged from the 1000 Genomes and Human Genome Diversity Project5, revealing the limitations of traditional groupings and capturing the complexities introduced by demographic events and evolutionary processes. This method embraces the context-specificity of genetic similarities that are salient depending on the question, markers of interest, and study individuals. Different numbers of communities are revealed depending on the resolution chosen and metric used, underscoring a fluid spectrum of genetic relationships and challenging the notion of universal categorization. We provide a web application (https://sohail-lab.shinyapps.io/GG-NC/) for interactive visualization and engagement with these intricate genetic landscapes.

show abstract

Topological stratification of continuous genetic variation in large biobanks

Cited by 3 publications

References 64 publications

Multi-ancestry polygenic risk scores using phylogenetic regularization

Multi-ancestry polygenic risk scores using phylogenetic regularization

Rare diseases load through the study of a regional population

The multi-scale complexity of human genetic variation beyond continental groups

Contact Info

Product

Resources

About