الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 134 |  |  |
| | | from | 134 | | |
Abstract
Computational models intelligently gather, filter, analyze and present health information to provide guidance to doctors for disease treatment based on detailed characteristics of each patient. The systems help to provide informed and consistent care of a patient as they transfer to appropriate hospital facilities and departments and receive various tests during their course of treatment. Bioinformatics is a field of computational science that has to do with the analysis of sequences of biological molecules. It usually refers to genes, DNA, RNA, or protein, and is particularly useful in comparing genes and other sequences in proteins and other sequences within an organism or between organisms, looking at evolutionary relationships between organisms, and using the patterns that exist across DNA and protein sequences to figure out what their function is.
In recent times, DNA microarray technology, which simultaneously measures the expressions of large numbers of genes and leads to high quality tumor identification, has become an advanced technology for cancer diagnosis and a very popular topic of research. Moreover, the advent of DNA microarray datasets has stimulated a new line of research in both bioinformatics and machine learning. Several metaheuristics algorithms and Machine Learning (ML) techniques are applied in DNA microarray technology for detecting normal and cancerous humans and classifying between different types of cancers.
Firstly, in this work, a new hybridization between Barnacles Mating Optimizer (BMO) algorithm, which is based on the intelligent swarm behaviors of barnacles during the mating process, and Support Vector Machines (SVM) called BMO-SVM is employed for a microarray gene expression profiling in order to select the most predictive and informative genes for cancer classification. BMO algorithm was proposed to solve optimization problems, known for efficient global searchability. A limited overview at the relevant literature shows that BMO has successfully solved challenging optimization problems in different fields. The statistical results revealed that the BMO algorithm can provide superior results in most tests compared with four well-known meta-heuristic optimization algorithms, such as Tunicate Swarm Algorithm (TSA), Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Artificial Bee Colony (ABC). Eventually, the BMO algorithm succeeded in achieving (i.e., 99.36\%), which is the highest percentage for the average of accuracy comparing with other algorithms. Moreover, the BMO algorithm achieves a high informational superiority percentage compared to other comparative algorithms.
Secondly, the Manta Ray Foraging Optimization (MRFO) algorithm, which is based on intelligent behaviors of manta rays for developing an efficient optimization technique to solve various optimization problems, is employed to select the most informative and predictive genes from microarray gene expression profile. MRFO is not used before for solving this problem and we tried to explore it ability in such domain, and the experimental results reveals that it is a good candidate in this area. The optimization results of MRFO outperforms the other meta-heuristic and conventional optimization algorithms. The employed algorithm is compared with some well-known selection algorithms, such as Harris hawks optimization (HHO), Black widow optimization(BWO), Grey Wolf Optimization(GWO), Genetic Algorithm(GA), Whale Optimization Algorithm(WOA), Particle Swarm Optimization(PSO), and Artificial Bee Colony(ABC). Experimental results on different microarray datasets showed the superiority of the employed algorithm in terms of classification accuracy and the least number of beneficial genes.