A novel gene selection method for gene expression data for the task of cancer type classification

N Özlem Özcan Şi̇mşek,Fikret Gürgen,Arzucan Özgür

doi:10.1186/s13062-020-00290-3

N Özlem Özcan Şi̇mşek, Fikret Gürgen + Show 1 more

Open Access

https://doi.org/10.1186/s13062-020-00290-3

Copy DOI

Journal: Biology Direct	Publication Date: Feb 8, 2021
Citations: 9	License type: open-access

Affiliation: Boğaziçi University

Abstract

Cancer is a poligenetic disease with each cancer type having a different mutation profile. Genomic data can be utilized to detect these profiles and to diagnose and differentiate cancer types. Variant calling provide mutation information. Gene expression data reveal the altered cell behaviour. The combination of the mutation and expression information can lead to accurate discrimination of different cancer types. In this study, we utilized and transferred the information of existing mutations for a novel gene selection method for gene expression data. We tested the proposed method in order to diagnose and differentiate cancer types. It is a disease specific method as both the mutations and expressions are filtered according to the selected cancer types. Our experiment results show that the proposed gene selection method leads to similar or improved performance metrics compared to classical feature selection methods and curated gene sets.

Highlights

Cancer is among the leading causes of death worldwide [1]
We propose a novel gene selection method targeting gene expression data for the task of cancer type classification
Each genomic disease occurs in consequence of a different mutation profile

Summary

Background

Cancer is among the leading causes of death worldwide [1]. It is a group of diseases and each cancer type is labeled by the primary area of the body where the cancer cells arise. Genomic tests reveal the gene mutations that may be driving a cancer’s behavior This information helps doctors while deciding on the patient’s personal treatment [2]. Whole genome sequences and variant calling are utilized for mutation analysis [3,4,5] Both coding and non-coding regions of the DNA are analyzed for the discovery of mutational signatures of cancer types. A number of studies have utilized gene expression data and addressed the classification of cancer types [6,7,8,9,10]. We propose a novel gene selection method targeting gene expression data for the task of cancer type classification.

Methods

Results and discussion

Conclusion