DNA Microarray Analysis Using Machine Learning to Recognize Cell Cycle Regulated Genes

Nowadays, automated computer tools and machine learning algorithms become compulsory. In fact, they facilitate analysis of complex and large datasets especially data related to the biological domain. One research area in bioinformatics is the DNA microarrays. This technology produces a large data volume. This data is often used to extract knowledge such as knowledge about genes expression. To this end, we propose a novel approach aiming to classify two types of genes. The first represents cell cycle regulated genes and the other represents the non-cell cycle ones. Approaches in state of art are based on processing the numerical data associated to the gene's evolution. In contrary, we try to use image data in order to be more efficient in the performed task. We use the classification algorithms SVM and KNN on our produced image data, related to each gene. The experiments show that our approach outperforms the state of art by a margin of 14 per cent.