An integrative gene selection with association analysis for microarray data classification

Abstract

The rising interest in integrative approach has shifted gene selection from purely data-centric to incorporating additional biological knowledge. Integrative gene selection is viewed as a promising approach in microarray data classification that took into consideration the complex relationships among genes. However, in most of the existing methods, the selection of genes is still based on expression values alone and biological knowledge is integrated at the end of analysis to verify experimental results or to gain biological insights. Thus, this paper proposed an integrative gene selection based on filter method and association analysis for selecting genes that are not only differentially expressed but also informative for classification. Association analysis is employed to integrate microarray data with multiple types of biological knowledge simultaneously, and to identify groups of genes that are frequently co-occurred in target samples. It has been tested on four cancer-related datasets, and two types of biological knowledge are incorporated, namely Gene Ontology (GO) and KEGG Pathways (KEGG). The experimental results show that the recommended GO based models, KEGG based models, and GO-KEGG based models outperformed the expression-only models by attaining better classification accuracies with lesser number of genes. The performance of the integrative models verified the efficiency and scalability of association analysis in mining microarray data.

Application of Biological Domain Knowledge Based Feature Selection on Gene Expression Data

“…However lately the gene selection process shifted from being purely data-centric to more incorporative analysis with additional biological knowledge. Integrative gene selection approaches incorporate domain knowledge from external biological resources during gene selection [9], [18], which improves interpretability and predictive performance. One of the widely used external ontology resources is the GO [19] which captures biological knowledge in a computable form that consists of a set of concepts and their relationships to each other.…”

“…Although the wrapping-based approaches can find the optimal set, yet it might be specific to the model used, such as SVM, decision trees or other models. In other words, it might be overfitting the data [18]. The main disadvantage of such methods are their difficulties for biological interpretation and they are unlikely to generate new biological knowledge.…”

“…Although the traditional gene selection approaches became popular for a long time, they have several drawbacks when one needs to precisely identify the underlying biological processes. Alternatively, integrative gene selection approaches incorporate domain knowledge from external biological resources during gene selection [9], [18], which improves interpretability and predictive performance. One of the widely used external ontology resources is the Gene Ontology (GO) [19] which provides i) cellular component (CC), ii) molecular function (MF), and iii) biological process (BP) terms for the products of each gene.…”

Integrative Gene Selection on Gene Expression Data: Providing Biological Context to Traditional Approaches

“…Integrative gene selection incorporates domain knowledge from external knowledge bases during its computations [4], [30]. Integrative gene selection leads to gene rankings that consider both the statistical significance of a gene in the dataset and the biological background information acquired through research.…”

“…Fang et al combined the Kyoto Encyclopedia of Genes and Genomes (KEGG) and GO with IG [30]. KEGG is a pathway knowledge base providing manually curated pathways [6].…”