MSalp-Epi: multi-objective salp optimisation for epistasis detection in genome-wide association studies
by S. Priya; R. Manavalan
International Journal of Intelligent Engineering Informatics (IJIEI), Vol. 10, No. 1, 2022

Abstract: Epistasis effects are depicted as interactions between different single nucleotide polymorphism (SNPs). It plays an essential role to recognise the individual susceptibility to complex human diseases. In this paper, we present a two-stage approach based on multi-objective salp optimisation for epistasis detection (MSalp-Epi) to detect two-locus epistasis associations for various simulated disease models. In the first stage, the salp optimisations use AIC and K2 scores as objective functions to find the non-dominated disease-related SNPs. In the second stage, the G-test statistic is applied over the non-dominated SNPs to attain the significant SNP pairs. The main objective of MSalp-Epi is to establish rapid and efficient multi-objective salp that accelerates the identification of disease-related SNP-SNP interactions from thousands of SNPs. The performance of MSalp-Epi is analysed and compared with MACOED and CSE. The outcome of the experimental analysis revealed that MSalp-Epi is superior to MACOED and CSE in terms of power, accuracy, true positive rate, (TPR) false detection rate Specificity, positive predicted value, F1-score and running time.

Online publication date: Thu, 30-Jun-2022

  Free full text Access Free full text access

If you still need assistance, please email subs@inderscience.com