Efficient Identification of Complex Diseases Through Epistasis Computational Models: A Review

R Manavalan; S Priya

Abstract

AbstractGenome-Wide Association Studies (GWAS) identify and characterize the genes that are associated with human diseases. One of the significant ongoing researches in GWAS is to identify the disease susceptible genes through Epistasis. The gene that masks the effects of other genes is called Epistasis. The gene interacts with another gene is known as epistatic or genetic interactions (GGIs). GWAS identifies the genetic variants of Single Nucleotide Polymorphism and also the interactions between SNPs to identify the disease susceptibility. The manual analysis of thousands of SNPs interactions was impractical to the physician. Hence, various statistical approaches and machine learning techniques were proposed to identify the genetic interactions in complex diseases such as Rheumatoid Arthritis (RA), Crohn diseases, Bipolar diseases (BD), Coronary Artery Disease and diabetes. This paper presents a survey on technological revolutions such as different data mining techniques, Machine Learning methods and statistical approaches used to identify the GGIs from Wellcome Trust Case Control Consortium (WTCCC) 7 diseases dataset, Crohn’s disease, RA, BD, Type I Diabetes (T1D) and Type II Diabetes (T2D) datasets. The issues behind the computational approaches to identify the diseases through Epistasis effects and also the parameters used by various researchers were analyzed. Keywords: Disease, epistasis, genes, genetic variations GWAS, interactions, SNPsCite this ArticleR Manavalan, S Priya. Efficient Identification of Complex Diseases Through Epistasis Computational Models: A Review. Research & Reviews: A Journal of Bioinformatics. 2020; 7(2): 21–32p.