Reduction of redundancy in diabetes-related gene categories utilizing conceptual similarity and gene expression information

Abstract

Insulin resistance or insulin deficiency can cause diabetes, which has a high level of glucose in the blood (American Diabetes Association 2014). It's one of the most pressing issues in modern public health. Around 366 million individuals had diabetes in 2011, according to the "International Diabetes Federation" (IDF). This figure is expected to rise to 552 million by the end of 2030. The average score of the semantic measure and the Pearson correlation were used to create a similarity index in this study. Gene similarity can be determined by this procedure. "One" indicates perfect similarity between two genes, whereas "zero" indicates perfect dissimilarity. Algorithms based on Pearson and semantic measures were used to identify genes that were genetically similar. PERL was used to write the algorithm (Practical Extraction and Reporting Language). PERL found gene-to-gene similarity and gradually deleted redundant information. An algorithm known as "greedy" was employed to remove all genes that above the 0.8 threshold value. An experiment using several datasets led to a final threshold value of 0.8.