Reaching the End-Game for GWAS: Machine Learning Approaches for the Prioritization of Complex Disease Loci

Abstract

Genome-wide association studies (GWAS) have revealed thousands of genetic loci that underpin the complex biology of many human traits. However, the strength of GWAS - the ability to detect genetic association by linkage disequilibrium (LD) - is also its limitation. Whilst the ever-increasing study size and improved design have augmented the power of GWAS to detect effects, differentiation of causal variants or genes from other highly correlated genes associated by LD remains the real challenge. This has severely hindered the biological insights and clinical translation of GWAS findings. Although thousands of disease susceptibility loci have been reported, causal genes at these loci remain elusive. Machine learning (ML) techniques offer an opportunity to dissect the heterogeneity of variant and gene signals in the post-GWAS analysis phase. ML models for GWAS prioritization vary greatly in their complexity, ranging from relatively simple logistic regression approaches to more complex ensemble models such as random forests and gradient boosting, as well as deep learning models, i.e., neural networks. Paired with functional validation, these methods show important promise for clinical translation, providing a strong evidence-based approach to direct post-GWAS research. However, as ML approaches continue to evolve to meet the challenge of causal gene identification, a critical assessment of the underlying methodologies and their applicability to the GWAS prioritization problem is needed. This review investigates the landscape of ML applications in three parts: selected models, input features, and output model performance, with a focus on prioritizations of complex disease associated loci. Overall, we explore the contributions ML has made towards reaching the GWAS end-game with consequent wide-ranging translational impact.

Keywords: artificial intelligence; candidate gene; clinical translation; data science; deep learning; genome-wide association study; genomics; machine learning.

FIGURE 1

Supervised Machine Learning Algorithm Training. (A) Data containing labeled genes (e.g., genes labeled as causal or non-causal for blood pressure – BP) and columns of features describing those genes are input into a machine learning algorithm. Machine learning algorithms firstly initialize themselves by applying their rules to a subset of the data (deemed training data) and its features at random. E.g., an algorithm’s first practice iteration can involve assigning feature importance at random (importance denoted by size of feature image). The algorithm uses its feature initialization to classify genes into either affecting BP (red genes) or not affecting BP (blue genes). Algorithms then use the practice predictions to calculate loss (an error rate) and iterate over the data again with applying the previous iteration’s loss calculation to adjust feature handling (B). With using the loss calculations the algorithm aims to improve predictive performance with each training iteration.

FIGURE 2

Supervised Machine Learning Models. Diagram detailing three machine learning model bases used in supervised learning, each providing varying algorithms most commonly used in post-GWAS prioritization.