AutoML for Multi-label Classification of piRNAs' Diseases

• Grant number: 25/03665-0
• Support Opportunities: Scholarships abroad - Research Internship - Scientific Initiation
• Start date: May 01, 2025
• End date: August 31, 2025
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Ricardo Cerri
• Grantee: Anna Carolina Brito Santos Farias
• Supervisor: Isaac Triguero Velazquez
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil
• Institution abroad: Universidad de Granada (UGR), Spain
• Associated to the scholarship: 24/09531-3 - Prediction of piRNAs Disease Associations Using Artificial Neural Networks, BP.IC

Abstract

PIWI-interacting RNAs (piRNAs) are small non-coding RNAs crucial for gene regulation, embryonic development, and cell differentiation. Recent studies highlight their involvement in diseases such as cancer, male infertility, Alzheimer's, and cardiovascular disorders. Despite their importance and the growing use of computational approaches in biology, predicting piRNA-associated diseases remains underexplored due to high data dimensionality and limited samples. The ongoing FAPESP project in Brazil applies neural networks to this problem, emphasizing the need for techniques like regularization and architecture optimization. In this context, AutoML offers a promising solution by automating model selection and hyperparameter tuning, enhancing robustness. However, comparative studies of AutoML tools for piRNA disease prediction are scarce. This BEPE project proposes evaluating multiple AutoML tools for multi-label classification and identifying their strengths and limitations. This approach will not only help identify the most effective tools for this task but also provide a better understanding of the advantages and limitations of each, contributing to the advancement of AutoML applications in biological studies.