Projects
The laboratory is dedicated to advancing research across diverse fields, including genomics, transcriptomics, metagenomics, proteomics, structural bioinformatics, and multi-omics analysis. In Computer Science, we focus on pioneering novel methodologies that leverage artificial intelligence, machine learning, heuristic search algorithms, and massively parallel processing architectures to address complex challenges in biological data analysis. In Biological Sciences, our research initiatives span a broad spectrum of applications, such as medical biotechnology, environmental biotechnology, animal biotechnology, microbial biotechnology, and plant biotechnology. The group is actively engaged in the following scientific-technical projects:
Computational Fluid Dynamics and Machine Learning for Sustainable Engineering In the realm of sustainable engineering, accurately simulating fluid dynamics presents numerous challenges due to their complex and nonlinear nature. Overcoming these challenges is crucial for driving scientific progress and enabling innovative engineering solutions. Read more
MAPSTROKE: a Computational Strategy to Improve Access to Acute Stroke Care This project aims to use computational strategies to identify optimal locations for new Acute Stroke Centers (ASCs), improving access to treatment, especially in rural areas of low and middle-income countries, thus addressing global disparities in stroke care. Read more
Machine Learning Interpretability Methods for Analysis of High-Dimensional Biological Data Biotechnology advancements rely on large-scale biological data from various technologies. This project aims to develop feature selection approaches in Machine Learning using interpretability strategies like Attention Mechanism and Layer-wise Relevance Propagation, along with multi-objective and self-adaptive optimization techniques. Read more
Data Science for Biotechnology Applications: solving large-scale challenges using explainable machine learning, metaheuristics, and high-performance computing This project develops bioinformatics tools using Machine Learning, heuristic search, and high-performance computing for human and animal health. It includes algorithms for adaptive metaheuristics, interpretable machine learning, and feature selection to analyze large, complex data effectively. Read more
Applying Machine Learning in the Identification and Validation of Biomarkers of Treatment Response and Cardiovascular Metabolism in Major Depressive Disorder Major Depressive Disorder (MDD) is a complex and multifactorial psychiatric condition. Given the high prevalence of MDD in patients with cardiovascular disease, a bidirectional causal relationship is plausible. We aim to investigate the impact of molecular mechanisms on susceptibility to MDD and cardiovascular disease and apply machine learning techniques to identify potential molecular markers. Read more
Traceability of Cannabis sativa L. in Brazilian Territory This project aims to use machine learning to identify genetic markers in Cannabis sativa L. from various regions of Brazil. By leveraging genomic analysis, including short repetitive sequences (SSR) and single nucleotide polymorphisms (SNPs), it seeks to understand the plant's biogeography and differentiate circulating genotypes based on geographic origin. Read more
Contribution of swine waste treatment processes in the control of antimicrobial resistance transmission for agropastoral systems This project maps antimicrobial resistance gene dissemination in Brazilian swine waste treatment systems, comparing cesspool and biodigester scenarios. Objectives include comparing microbiomes, identifying resistome markers, and utilizing machine learning to trace AMR profiles, informing policy and enhancing surveillance. Read more
AICaBI: Artificial Intelligence for Cancer Biomarkes Identification This interdisciplinary project aims to enhance personalized medicine by integrating computational methods and strategies in a Bioinformatic context. Focusing on cancer prevention, diagnosis, and prognosis precision, it seeks to answer whether meta-heuristics, machine learning, and biological annotation data can deliver high-quality solutions for identifying cancer biomarkers. Read more
Mesoscopic Molecular Dynamics Simulations: Development of Models and Computational Strategies for Complex Structural Bioinformatics Problems This project merges microscopic and macroscopic effects with a Mesoscopic Molecular Dynamics (MMD) model. Integrating Molecular Dynamics (MD) precision with Computational Fluid Dynamics (CFD) scalability via the Lattice Boltzmann Method (LBM), it addresses crucial microscopic effects. Brazil and Germany collaborate to offer interdisciplinary research and education to PhD students and researchers. Read more
Parallel and Distributed Metaheuristics for Structural Bioinformatics This international collaboration aims to develop metaheuristic models using robust parallel and distributed computing for structural bioinformatics. Partnering institutions include UFRGS (Brazil), UNSL (Argentina), USACH (Chile), and UPMC (France), addressing unanswered questions in structural bioinformatics beyond traditional approaches. Read more
Multi-modal and multi-objective heuristic search methods for Structural Bioinformatics problems This project develops metaheuristic approaches for complex problems in Structural Bioinformatics, such as predicting macromolecular structures and molecular docking for drug development. Metaheuristics offer efficient solutions when exact methods are impractical, aiming to enhance current strategies in the field. Read more