Hierarchical classification of transposable elements and protein functions making use of machine learning

• Grant number: 16/25078-0
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: March 01, 2017
• End date: June 30, 2018
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Ricardo Cerri
• Grantee: Bruna Zamith Santos
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil
• Associated scholarship(s): 17/13218-5 - Predicting protein functions via interaction prediction, BE.EP.IC

Abstract

Proteins are macromolecules responsible for almost every task necessary for the maintenance of cells, playing an essential role in the behavior and regulation of organisms. Advances in Molecular Biology have allowed an almost complete listing of the proteins that make up the organisms. However, there are a large number of proteins whose function is still unknown, opening space for a new research focus in Molecular Biology. In parallel, Transposable Elements (TEs) are DNA sequences that have the ability to change their location within the genome, affecting, thereby, the activity of certain genes. When TEs are inserted in other genes, they can alter or reduce the activity of certain proteins. Given the need for correct identification and classification of TEs and protein functions, there are several techniques currently employed to achieve this goal. The main ones involve a lot of manual work or they are very specific, restricting the set of data to be classified. Moreover, the classes involved in these problems are structured hierarchically, a fact often ignored by the proposed techniques. Thus, this project proposes the use of Machine Learning (ML) for the hierarchical classification of TEs and functions of proteins. In addition to the use of several methods already known, a new contribution will be the investigation of different ways of using positive and negative examples during the induction of hierarchical methods. This is especially important in methods that follow the top-down strategy. All methods will be evaluated using specific measures for hierarchical problems. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

PEREIRA, GEAN TRINDADE; SANTOS, BRUNA ZAMITH; CERRI, RICARDO; IEEE. A Genetic Algorithm for Transposable Elements Hierarchical Classification Rule Induction. 2018 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION (CEC), v. N/A, p. 8-pg., 2018-01-01. (15/14300-1, 16/25078-0)

SANTOS, BRUNA Z.; NAKANO, FELIPE K.; CERRI, RICARDO; VENS, CELINE; HUTTER, F; KERSTING, K; LIJFFIJT, J; VALERA, I. Predictive Bi-clustering Trees for Hierarchical Multi-label Classification. MACHINE LEARNING AND KNOWLEDGE DISCOVERY IN DATABASES, ECML PKDD 2020, PT III, v. 12459, p. 18-pg., 2021-01-01. (16/25078-0, 17/13218-5)