Deep tree-ensembles for predicting lncRNA-associated diseases

• Grant number: 24/05438-9
• Support Opportunities: Scholarships abroad - Research Internship - Scientific Initiation
• Start date: July 01, 2024
• End date: October 31, 2024
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Ricardo Cerri
• Grantee: Lívia Umberto Bertoni
• Supervisor: Celine Vens
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil
• Institution abroad: University of Leuven, Kulak Kortrijk (KU Leuven), Belgium
• Associated to the scholarship: 22/14762-9 - Multilabel classification machine learning method for predicting long non coding RNA-associated diseases, BP.IC

Abstract

Long non-coding RNAs (lncRNAs) are RNA sequences longer than 200 nucleotides that are not translated into functional proteins. LncRNAs play a crucial role in several vital activities; therefore, dysfunctions of lncRNAs are associated with a wide range of diseases. Thus, is important to identify and detect lncRNAs, essencials to disease diagnosis and therapy. Given the adversities involving biotechnological approaches, machine learning arises as an outstanding alternative to perform disease prediction based on lncRNA data. LncRNAs are associated with at least one disease. Many machine learning computational models have been proposed for lncRNA-disease association prediction. However, experimental results indicate that various methods exhibited suboptimal performance. In this setting, a new deep tree ensemble method (DTE) stands out with superior prediction results, which integrates a representational learning component. This project proposes the application of this DTE model for the lncRNA-disease association prediction task. More precisely, we will investigate the performance of the method for this purpose, as well as evaluate its prediction performance against state-of-the-art methods.