Recovering Direct Effects in Genetics: A Comparison

TL;DR

This paper compares methods for identifying directly disease-related SNPs in genetics, highlighting the effectiveness of stability selection and direct effect testing across various correlation scenarios.

Contribution

It provides a comprehensive simulation-based comparison of existing methods for true sparsity recovery in genetic data analysis.

Findings

Lasso performs well with low predictor correlation.

Specialized methods are needed for high correlation.

Stability selection and direct effect testing are robust across scenarios.

Abstract

In genetics it is often of interest to discover single nucleotide polymorphisms (SNPs) that are directly related to a disease, rather than just being associated with it. Few methods exist, however, addressing this so-called `true sparsity recovery' issue. In a thorough simulation study, we show that for moderate or low correlation between predictors, lasso-based methods perform well at true sparsity recovery, despite not being specifically designed for this purpose. For large correlations, however, more specialised methods are needed. Stability selection and direct effect testing perform well in all situations, including when the correlation is large.