Ant colony algorithm for analysis of gene interaction in high-dimensional association data

Abstract

-In recent years there has been much focus on the use of single nucleotide polymorphism (SNP) fine genome mapping to identify causative mutations for traits of interest; however, many studies focus only on the marginal effects of markers, ignoring potential gene interactions. Simulation studies have show that this approach may not be powerful enough to detect important loci when gene interactions are present. While several studies have examined potential gene interaction, they tend to focus on a small number of SNP markers. Given the prohibitive computation cost of modeling interactions in studies involving a large number SNP, methods need to be develop that can account for potential gene interactions in a computationally efficient manner. This study adopts a machine learning approach by adapting the ant colony optimization algorithm (ACA), coupled with logistic regression on haplotypes and genotypes, for association studies involving large numbers of SNP markers. The proposed method is compared to haplotype analysis, implemented using a sliding window (SW/H), and single locus genotype association (RG). Each algorithm was evaluated using a binary trait simulated using an epistatic model and HapMap ENCODE genotype data. Results show that the ACA outperformed SW/H and RG under all simulation scenarios, yielding substantial increases in power to detect genomic regions associated with the simulated trait.Key Words: genome, simulation, SNP Algoritmo colônia de formigas para análise de interação gênica em dados de associação de alta dimensão RESUMO -Nos últimos anos muita atenção tem sido dada ao uso de polimorfismos de nucleotídeos simples (SNP) para mapeamento fino do genoma, visando identificar mutações efetivas em características de interesse; todavia, muitos estudos focam apenas os efeitos marginais dos marcadores, ignorando as potenciais interações entre genes. Estudos de simulação tem mostrado que esta abordagem pode não ser poderosa o suficiente para detectar loci importantes quando interações entre genes estão presentes. Vários estudos tem examinado potenciais interações gênicas, porém focando um pequeno número de marcadores SNP. Devido ao proibitivo custo computacional para modelar interações em estudos envolvendo um grande número de SNP's, precisam ser desenvolvidos métodos que considerem potenciais interações gênicas, de uma forma computacionalmente eficiente. Este estudo adota a abordagem de um mecanismo de aprendizagem, adaptando o algoritmo de otimização colônia de formigas (ACA), combinado com regressão logística em função dos haplótipos e genótipos, para estudos de associação envolvendo grande número de marcadores SNP. O método proposto é comparado à análise de haplótipos, implementado usando uma janela deslizante (SW/H), e a associação de genótipos de lócus único (RG). Cada algoritmo foi avaliado usando uma característica binária simulada usando um modelo epistático e dados genotípicos do HapMap ENCODE. Os resultados mostram que o ACA superou o SW/H e RG em todos os cenários de simulação, pr...

Estimation of adjustment factors for 3X milking

“…Estimates of residual variance were also higher at the beginning and at the end of lactation. Similar results for residual variance estimates were also reported by Rekaya et al [84] and by Swalve [97].…”

A Review of Ant Colony Optimization Based Methods for Detecting Epistatic Interactions

“…Rekaya et al [31], [32] provided a probability function with two-layer pheromones, which was defined as…”

“…Uppu et al [11] reviewed 7 groups of epistasis detection methods, including exhaustive search methods, random forests, neural networks, support vector machines, regression models, Bayesian approaches, and ACO approaches. For the group of ACO approaches, they provided a brief overview of 7 extensions and modifications of the generic ACO algorithm for detecting epistatic interactions [26], [31], [35], [37], [40], [43], [46].…”

A Survey on Swarm Intelligence Search Methods Dedicated to Detection of High-Order SNP Interactions

“…To analyze epistasis in human disease, Greene et al combined expert knowledge and multifactor dimensionality reduction (MDR) with the ACO algorithm to quickly explore the epistatic interactions, and the expert knowledge was obtained from tuned relief (TuRF) [25]. Rekaya and Robbins employed an ACO algorithm (ACA) to analyze gene interactions; they used two-layer pheromones for ants to choose a path and logistic regression to evaluate the association between genotype and haplotype [26]. Wang et al proposed a two-stage ACO algorithm (AntEpiSeeker) that aims to discover a highly suspected and reduced SNP set quickly in the first stage.…”