Recognition of cancer mediating biomarkers using rough approximations enabled intuitionistic fuzzy soft sets based similarity measure

In recent years, gene expression analysis has become crucial in studying microarray and provides several techniques related to computational biology and bioinformatics. This article introduces a novel Intuitionistic fuzzy set under rough multigranulation approximation based feature extraction from microarray gene expression dataset. The proposed model identifies cancer mediating human biomarkers using an Intuitionistic fuzzy soft set based similarity measure. Firstly, Intuitionistic fuzzy rough reduct is devised in the microarray datasets preceded by an entropy based pre-processing step. Intuitionistic fuzzy rough reduct produces the feature sets from the microarray datasets through roughness and accuracy measures that are differentially expressed in a cancerous state. Rough multigranulation approximations reduce the dimensions of the microarray datasets. Thereafter, a weighted similarity measure is designed using the intuitionistic fuzzy soft set for the classification of biomarkers having significant expression patterns from the normal state to the carcinogenic state. The proposed method is demonstrated on six microarray datasets and is validated with different performance metrics resulting in better effectiveness than the state-of-the-art methods. (c) 2022 Elsevier B.V. All rights reserved.

Automated summary

Gene expression analysis plays a crucial role in microarray research. This article introduces a novel feature extraction method based on Intuitionistic fuzzy set for identifying cancer-related human biomarkers. The experimental results demonstrate the effectiveness of this method on microarray datasets.