(CDRGI)-Cancer detection through relevant genes identification

Feras Al-Obeidat
Álvaro Rocha
Maryam Akram
Saad Razzaq
Fahad Maqbool

Document Type

Article

Source of Publication

Neural Computing and Applications

Publication Date

2-5-2021

Abstract

Cancer is a genetic disease that is categorized among the most lethal and belligerent diseases. An early staging of the disease can reduce the high mortality rate associated with cancer. The advancement in high throughput sequencing technology and the implementation of several Machine Learning algorithms have led to significant progress in Oncogenomics over the past few decades. Oncogenomics uses RNA sequencing and gene expression profiling for the identification of cancer-related genes. The high dimensionality of RNA sequencing data makes it a complex and large-scale optimization problem. CDRGI presents a Discrete Filtering technique based on a Binary Artificial Bee Colony coupling Support Vector Machine and a two-stage cascading classifier to identify relevant genes and detect cancer using RNA seq data. The proposed approach has been tested for seven different cancers, including Breast Cancer, Stomach Cancer (STAD), Colon Cancer (COAD), Liver Cancer, Lung Cancer (LUSC), Kidney Cancer (KIRC), and Skin Cancer. The results revealed that the CDRGI performs better for feature reduction while achieving better classification accuracy for STAD, COAD, LUSC and KIRC cancer types.

ISSN

0941-0643

Publisher

Springer Nature

Disciplines

Medicine and Health Sciences

Keywords

Support vector machine, Cascading classifier, Discrete filtering, Artificial bee colony, Gene expression, CatBoost classifier, Convolutional neural network

Scopus ID

85100525522

Indexed in Scopus

yes

Open Access

no

Link to Full Text

Share

COinS