Title: Machine learning approaches for disease genes prediction

Authors: Priya Sadana; Isha Kansal; Vikas Khullar

Addresses: Chitkara University Institute of Engineering and Technology, Chitkara University, Punjab, India ' Chitkara University Institute of Engineering and Technology, Chitkara University, Punjab, India ' Chitkara University Institute of Engineering and Technology, Chitkara University, Punjab, India

Abstract: The identification of genes involved in human hereditary diseases frequently necessitates the examination of a large number of potential candidate genes, which can be time-consuming and expensive. Genome-wide techniques such as association studies and linkage analysis frequently select many hundreds of positional candidates. Earlier binary classification methods used disease-causing and healthy genes as positive and negative training sets but risked including unknown disease genes. This work aims to discuss machine learning-based methods for disease susceptibility gene identification. Recent advancements, include complex methods like ensemble and deep learning. Then, we evaluated several well-known machine learning-based disease gene prediction algorithms. We concluded by discussing the pros and cons of different methods and their interpretability and reliability. A comparative study demonstrates the effectiveness of proposed approaches, contributing to the advancement of disease gene identification methodologies while highlighting their interpretability and reliability.

Keywords: neurological disorder; gene prediction; binary classification; semi supervised learning; SSL.

DOI: 10.1504/IJDMB.2026.154701

International Journal of Data Mining and Bioinformatics, 2026 Vol.30 No.3/4, pp.342 - 360

Received: 04 Nov 2023
Accepted: 15 May 2024
Published online: 10 Jul 2026 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article