Grid-based stochastic search for hierarchical gene-gene interactions in population-based genetic studies of common human diseases

Abstract

BackgroundLarge-scale genetic studies of common human diseases have focused almost exclusively on the independent main effects of single-nucleotide polymorphisms (SNPs) on disease susceptibility. These studies have had some success, but much of the genetic architecture of common disease remains unexplained. Attention is now turning to detecting SNPs that impact disease susceptibility in the context of other genetic factors and environmental exposures. These context-dependent genetic effects can manifest themselves as non-additive interactions, which are more challenging to model using parametric statistical approaches. The dimensionality that results from a multitude of genotype combinations, which results from considering many SNPs simultaneously, renders these approaches underpowered. We previously developed the multifactor dimensionality reduction (MDR) approach as a nonparametric and genetic model-free machine learning alternative. Approaches such as MDR can improve the power to detect gene-gene interactions but are limited in their ability to exhaustively consider SNP combinations in genome-wide association studies (GWAS), due to the combinatorial explosion of the search space. We introduce here a stochastic search algorithm called Crush for the application of MDR to modeling high-order gene-gene interactions in genome-wide data. The Crush-MDR approach uses expert knowledge to guide probabilistic searches within a framework that capitalizes on the use of biological knowledge to filter gene sets prior to analysis. Here we evaluated the ability of Crush-MDR to detect hierarchical sets of interacting SNPs using a biology-based simulation strategy that assumes non-additive interactions within genes and additivity in genetic effects between sets of genes within a biochemical pathway.ResultsWe show that Crush-MDR is able to identify genetic effects at the gene or pathway level significantly better than a baseline random search with the same number of model evaluations. We then applied the same methodology to a GWAS for Alzheimer’s disease and showed base level validation that Crush-MDR was able to identify a set of interacting genes with biological ties to Alzheimer’s disease.ConclusionsWe discuss the role of stochastic search and cloud computing for detecting complex genetic effects in genome-wide data.

Machine Learning Approaches and Applications in Genome Wide Association Study for Alzheimer’s Disease: A Systematic Review

“…These variations were incorporated of a model that includes Crush, a stochastic search technique to explore relations between genes in genome-wide data as an application of multifactor dimensionality reduction (MDR). Applying the approach to an AZ GWAS dataset, results showed that Crush-MDR was capable of identifying a collection of interacting genes with biological linkages to Alzheimer's disease [57]. Xu et al employed a Time-Varying Group Sparse Additive Model (TV-GroupSpAM) to carry out a two-stage SNP selection.…”

Epistatic Features and Machine Learning Improve Alzheimer’s Risk Prediction Over Polygenic Risk Scores

“…Epistatic interaction terms were selected using multifactor dimensionality reduction (MDR) [19], a nonparametric approach that collapses the genotype combinations into high risk or low risk, then tests this new variable's association with the phenotype using cross validation. The Crush-MDR algorithm [13] uses an evolutionary algorithm guided by expert knowledge to mine the space of SNP interactions. Candidate SNPs to include in the interaction mining were selected within each training set of unrelated individuals.…”

The Application of Artificial Intelligence in the Genetic Study of Alzheimer’s Disease

“…Expert knowledge from databases (such as gene ontology) or literature sources (such as PubMed) was used to filter gene datasets before the analysis. They applied this method to the GWAS dataset from ADNI and identified a set of interacting genes related to AD [ 137 ].…”