Research
Research interests and projects
My work focuses on the development and evaluation of computational methods for complex biomedical data. I design and analyze machine learning methodologies tailored to high-dimensional molecular datasets and clinical records, with applications in disease modeling and understanding, biomarker discovery, and clinical prediction.
I am particularly interested in methodological challenges such as model robustness and fairness, feature selection, representation learning on graphs, and the reliable evaluation of predictive systems in healthcare contexts.
I have worked with a variety of problems within the fields of Medicine and Genomics. Recent projects are related (but not limited) to the following topics:
- Computational inference and analysis of gene regulatory networks
- Integration of multi-omics data to study complex human diseases
- Machine learning models in clinical and biological data science
- Metanalysis and feature selection techniques to identify disease-related genes
- Network-based approaches to investigate diseases mechanisms and biomarkers
- Prediction and analysis of microRNAs' target genes
On the computational side, my interests are mainly on:
- Supervised, unsupervised and semi-supervised learning
- Machine learning on graphs
- Feature selection methods
- Machine learning with multi-view data
- Ensemble learning and the 'wisdom of crowds' theory
All students under my supervision are members of the Machine Learning & Applications in Biomedical Data Lab (MLAB), based at INF/UFRGS. MLAB is a research group dedicated to advancing machine learning methodologies and their application to diverse biomedical data, including clinical, epidemiological, laboratory, omics, and imaging datasets. Our work addresses both computational challenges and the biological and clinical relevance of the models we develop. We value interdisciplinary collaboration and welcome discussions on research problems in healthcare and genomics that require rigorous and innovative computational approaches.
If you would like to propose a collaboration do not hesitate to contact me.
Projects
Below I highlight some ongoing and recent projects. For a complete relation of research projects, please see my CV at Lattes Platform.
- 2022 - present. Aprendizado de máquina com dados heterogêneos para modelos preditivos mais precisos em saúde e genômica. Coordination: Profa. Mariana Recamonde Mendoza. Funded by CNPq.
- 2022 - 2025. MARCS: Modelos de Aprendizado de máquina Robustos e Confiáveis para a Saúde. Coordination: Profa. Mariana Recamonde Mendoza. Funded by FAPERGS.
- 2023 - 2024. Construindo a inteligência da ciência cidadã para estratégias de preparação e resposta a pandemias: um estudo piloto. Coordination: Profa. Mariana Recamonde Mendoza. In collaboration with HCPA. Funded by I-DAIR (now HealthAI).
- 2020 - 2023. Diagnóstico e predição clínica em oncologia: análise integrativa de dados ômicos através de modelos complexos de aprendizado de máquina. Coordination: Profa. Mariana Recamonde Mendoza. In collaboration with HCPA. Funded by CNPq/AWS.
- 2020 - 2023. CIDIA-19 - CIência de Dados e Inteligência Artificial para combater a COVID-19. Coordinator: Prof. João Comba. In collaboration with HCPA and HMV. Funded by FAPERGS.