Predicting Critical Genes from Genomic Data Using Artificial Gorilla Troops Optimizer

EasyChair Preprint 7428

Abstract

The enhancement of microarray technology motivates researchers in computational biology to apply computational methods to biological data. A genomic dataset contains more than 300,000 genes. Among them, only a few are responsible for disease. Predicting critical genes is hard to do in this era. A meta-heuristic optimization algorithm can play a significant role in it. Several gene selection techniques for disease classification work for meta-heuristic algorithms that can give good classification accuracy with fewer selected genes. A meta-heuristic algorithm can play a significant role in predicting critical genes. This technique plays a vital role in optimization techniques, which are easy to transform and do not require any derivatives. This technique can exceed the local minimums by accepting bad moves as solutions. The Artificial Gorilla Troops Optimizer is a nature-inspired population-based met-heuristic-based algorithm that is inspired by the collective behaviour of gorillas. In this work, the performance of AGTO and MGTO is tested on 25 selected benchmark test functions to check their superiority. AGTO and MGTO are compared by their best objective value. Also, the running time of the GTO algorithm on MATLAB and Python code is compared across 25 selected benchmark functions. And also, execution time is compared between MATLAB and PYTHON. The results prove that GTO is superior to other metaheuristics algorithms, but MGTO gives better results than AGTO. As AGTO and MGTO are superior or more competitive than other met-heuristic algorithms, they will easily predict critical genes from genomic datasets with high classification accuracy.

Keyphrases: Artificial Gorilla Troops Optimizer, Gene selection, Meta-heuristic, Modified Gorilla Troops Optimizer, Optimization

@booklet{EasyChair:7428,
  author    = {Rukhsun Ara Parvin and Biswajit Jana and Sriyankar Acharyya},
  title     = {Predicting Critical Genes from Genomic Data Using Artificial Gorilla Troops Optimizer},
  howpublished = {EasyChair Preprint 7428},
  year      = {EasyChair, 2022}}