Benchmarking of univariate pleiotropy detection methods applied to epilepsy

Area of Science:

• Genetics
• Statistical Genetics
• Genomic Epidemiology

Background:

• Pleiotropy, where one gene influences multiple traits, is common and offers insights into disease etiology.
• Genome-wide association studies (GWAS) often analyze single traits, limiting pleiotropy detection to univariate methods with unknown comparative performance.
• Evaluating univariate pleiotropy detection methods is crucial for advancing genetic research.

Purpose of the Study:

• To compare the performance of five univariate pleiotropy detection methods: meta-analysis, ASSET, conditional false discovery rate (cFDR), cross-phenotype Bayes (CPBayes), and pleiotropic analysis under the composite null hypothesis (PLACO).
• To identify the most effective method balancing power and false positive rates.
• To apply the best-performing method to identify pleiotropic genetic loci in complex epilepsies.

Main Methods:

• Extensive computer simulations were conducted, varying pleiotropy models, number of causal variants, trait overlap, effect sizes, trait prevalence, and sample sizes.
• Five univariate pleiotropy detection methods were evaluated based on their ability to detect pleiotropy.
• The ASSET method was applied to a dataset from the International League Against Epilepsy (ILAE) consortium for genetic generalized epilepsy and focal epilepsy.

Main Results:

• The ASSET method demonstrated the optimal balance between statistical power and control of false positives across various simulated scenarios.
• Application of ASSET to ILAE data identified a novel candidate pleiotropic locus at 17q21.32.
• The known pleiotropic locus 2q24.3 was confirmed, and functional analyses supported the novel 17q21.32 finding.

Conclusions:

• ASSET is a robust method for detecting univariate pleiotropy, offering a good trade-off between power and specificity.
• The study identified 17q21.32 as a novel candidate locus for pleiotropy in complex epilepsies, warranting further investigation.
• This research enhances the understanding of pleiotropy in complex diseases and provides a valuable tool for genetic discovery.