Improvement of the target selection process in transcriptomics data

The knowledge of processes related to protein activation is directly related to a cascade of genes activation and inhibition. The study of gene expression allows to select differentially expressed genes potentially involved in a similar process. However, some genes with a common biological characteristics are not selected in terms of expression. The use of gene ontology which is a reliable source of information on genes could conduct and improve the gene selection process. In this paper, we propose a new approach which combines the coexpression of significantly and non-significantly expressed genes with their ontological information to improve the gene selection process in transcriptomics data.