Modelling RNA-seq read counts by grey relational analysis

Abstract

This paper proposes a feature selection approach for RNA-seq read counts modelling based on grey relational analysis (GRA). Read counts are transformed to microarray-like data to facilitate normal-based statistical methods. GRA is designed to select differentially expressed genes by integrating outcomes of five individual feature selection methods including two-sample t-test, entropy test, Bhattacharyya distance, Wilcoxon test and receiver operating characteristic curve. GRA performs as an aggregate filter method through combining advantages of the individual methods to produce significant feature subsets that are then fed into classifiers for evaluation. The proposed approach is verified by using two benchmark real datasets and the five-fold cross-validation method. Experimental results show the performance dominance of the GRA-based feature selection method against its competing methods. This implies that the proposed method can be implemented effectively in real practice for medical applications such as disease diagnosis using RNA-seq data analysis.