Fast and Accurate Exhaustive Higher-Order Epistasis Search with BitEpi

Abstract

Motivation: Higher-order epistatic interactions can be the driver for complex genetic diseases. An exhaustive search is the most accurate method for identifying interactive SNPs. While there is a fast bitwise algorithm for pairwise exhaustive searching (BOOST), higher-order exhaustive searching has yet to be efficiently optimized. Results: In this paper, we introduce BitEpi, a program to detect and visualize higher-order epistatic interactions using an exhaustive search. BitEpi introduces a novel bitwise algorithm that can perform higher-order analysis more quickly and is the first bitwise algorithm to search for 4-SNP interactions. Furthermore, BitEpi increases detection accuracy by using a novel entropy-based power analysis. BitEpi visualizes significant interactions in a publication-ready interactive graph. BitEpi is 56 times faster than MDR for 4-SNP searching and is up to 1.33 and 2.09 times more accurate than BOOST and MPI3SNP respectively.

Novel Alzheimer’s disease genes and epistasis identified using machine learning GWAS platform

“…22 University Of Pittsburgh, Pennsylvania, USA. 23 Cornell University, New York, USA. 24 Albert Einstein College of Medicine of Yeshiva University, New York, USA.…”

“…Using a novel false discovery rate (FDR) method [22], we are able to use VariantSpark's random-forest-based feature selection approach to narrow down the genome-wide search space to the subset of variants enriched with epistatic interactions. We then apply BitEpi [23] to perform an exhaustive search of this subset to annotate pairwise and higher-order, statistically significant interactions between the variants. We also explore the proportion of phenotypic variance captured by VariantSpark versus the traditional logistic regression (LR) methods.…”

Identification of regulatory SNPs and epistatic SNP pairs using deep learning and information theory

“…This approach allows to quickly determine the frequencies of the genotype combinations and whether two samples have the same combination, which is necessary to calculate MI. Similar approaches have been used in various other methods for epistasis detection [81,166,213]. As shown in [88,159,214], the value of the mutual information and its possible range is strongly dependent on the alphabet size and the marginal distributions of the variables.…”

“…Instead, these methods and MIDESP simply report a certain percentage or absolute number of the top results with the highest MI-values [76,80,301]. As an alternative approach, several methods suggest to apply permutation testing in order to obtain p-values [76,78,79,213]. However, in order to apply permutation testing to MIDESP, it would become necessary to re-estimate the expected background level of each SNP for each round of permutation.…”