From cells to social interactions: an AI-based integrative approach to unveil the underlying mechanisms of human behaviors and phenotypes

• Grant number: 23/18337-3
• Support Opportunities: Research Projects - Thematic Grants
• Start date: June 01, 2025
• End date: May 31, 2030
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: André Fujita
• Grantee: André Fujita
• Host Institution: Instituto de Matemática e Estatística (IME). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Pesquisadores principais: Helder Takashi Imoto Nakaya ; João Ricardo Sato
• Associated researchers: Alexandre Alarcon Steiner ; Ary Gadelha de Alencar Araripe Neto ; Bruno de Bezerril Andrade ; Daniel Oliveira Dantas ; Daniel Yasumasa Takahashi ; Eduardo Lani Volpe da Silveira ; Frederico Moraes Ferreira ; Gabriel Nassar Reich Goldstein ; José Deney Alves Araujo ; Katerina Lukasova ; Marcos Inácio Severo de Almeida ; Mauro César Cafundó de Morais ; Patricia Severino ; Suellen Mary Marinho dos Santos Andrade ; Vanderson de Souza Sampaio

Abstract

Handling and interpreting large volumes of multimodal data is a central challenge, requiring specialized AI and computational tools. This project will employ high-performance computing and AI algorithms on data obtained from cutting-edge technologies to investigate human behaviors and phenotypes at various biological scales, from molecular to environmental levels. It capitalizes on our previous FAPESP thematic project, which focused on data integration through network science and produced over 100 scientific papers and computational tools. Four key subprojects anchor the research. The first focuses on gene expression and phenotype relationships using spatial transcriptomics, employing AI to process this complex data. The second subproject uses AI alongside fNIRS and eye-tracking to study brain activity in educational contexts, aiming to understand learning processes. The third examines physiological responses during social interactions, using heart rate variability as a measure. The final subproject involves creating an AI-based tool for integrating diverse data sets, including Omics, brain activity, and environmental factors. The AI approach for managing, processing, and interpreting large-scale, multimodal datasets is central to all sub-projects. This approach will address not only the volume and complexity of the data but also reveal insights into the interconnectedness of biological systems. The project seeks a budget of 4.4 million Reais to fund technological resources and skilled personnel, including acquiring cutting-edge equipment and a team specialized in AI algorithm implementation and data analysis. (AU)