Vol. 8, Issue 7, Part E (2022)

Abstract
The Microarray data consists of various gene expression profiles which are used to identify disease, but the huge dimensions of the data make processing difficult. Assorted techniques like Artificial Bee colony are used to reduce the data and make its processing effective. The gene expression level is used to recognize diseases and help in their treatment. By identifying the early onset of disease, this technique will be able to arrest its further progress and control its prognosis, to a certain extent, through facilitating the development of appropriate medication using DNA datasets. The leukaemia dataset obtained the best accuracy at 81.667% with 35 genes in trial 3. The breast cancer dataset has the best accuracy of 70% with 20 genes in trial 1, and the prostate cancer dataset has its best accuracy at 85% with 35 genes in trial 2. Our proposed work focuses on the selection of affordable set of gene for a disease with an acceptable level of accuracy from the microarray dataset that is efficiently used to identify the cancer-causing genes in particular, or other disease-causing genes in general, in patients.