Paper Title
AN IMPROVED ARTIFICIAL RABBIT’S OPTIMIZATION (ARO) ALGORITHM FOR IDENTIFYING MUTATED DRIVER GENES IN CANCER

Abstract
Abstract - The struggle of medics, computational biologist and experts in bioinformatics to find a cure for cancer is one of the most difficult problems in the world today. Given the large amounts of genomic data that is generated on a daily basis, it is becoming increasing difficult to evaluate and investigate this data. This is due to the fact that cancer data is heterogeneous, consisting of passenger genes which do not contribute to oncogenesis as well as driver genes which are directly led to oncogenesis. Hence identifying these driver genes from passenger genes in these large chunks of data increasingly becomes a difficult task. Considering the previous methods that have been developed to solve this problem, in this research we propose a bio-inspired method called Artificial Rabbit’s Optimization (ARO) that integrates a mutation phase to be used to solve this problem of identifying cancer driver genes. This method merges the survival behavior of rabbits through exploration and exploitation to handle both global and local search respectively, with a gene interaction network to improve the accuracy of discovering cancer driver genes. The model is applied to 4 different types of cancers: breast cancer, brain cancer, prostate cancer and ovarian cancer. The results demonstrate that the proposed model can identify well-known labelled canonical driver genes while prioritizing them over unknown cancer driver genes. GBM found 9 genes, BRCA found 25 genes, OV found 4 genes and PRAD found 12 genes in the top 30 ranked genes as recognized by the NCG7.0. Keywords - Cancer, Driver Genes, Artificial Rabbits Optimization, Metaheuristic, Biological Interactions.