My research interests centre on the application of AI techniques to the study of proteins and their interactions.
Research topics and software developed
PSICO is an algorithm that builds a protein structure by narrowing down the volumes where atoms can be placed. The purpose is to process structural Nuclear Magnetic Resonance data, and to complement it with structural constraints from other sources such as homology modelling or secondary structure prediction.
BiGGER is an algorithm for searching possible configurations of protein complexes and finding the most likely models for the interaction. It relies on constraint programming techniques for an efficient search and to use relevant data from any experimental technique that can provide information about the interaction.
Modelling and Computation.
Chemera is an application that serves as interface for PSICO and BiGGER and also includes a set of tools to visualise and manipulate structures, integrate information from remote services and compute some electrostatic properties of proteins.
CREAM is an application for modelling electrochemistry and reaction kinetics.
Research Projects (as principal investigator)
CREMA (2011-12): Constrained Refining of Multiple Alignments
Research Projects (as a participant)
Structural characterization of mRNA localization mechanisms (2008-11)
This project focuses on protein-mRNA interaction, protein-protein interactions for RNP transport and mRNA as regulators of protein expression. It was funded by the Fundação para a Ciência e Tecnologia, MCTES.
Mechanisms of Electron Transfer in Metalloproteins (2005-09)
This project focused on the voltametric characterization of electron transfer in metalloproteins. It was funded by the Fundação para a Ciência e Tecnologia, MCTES.
REWERSE(2004-08):Reasoning on the Web with Rules and Semantics
REWERSE was an international "Network of Excelence" funded by the EU Comission and Switzerland for the development of reasoning languages for a semantic web. I participated in the A2 (bioinformatics) group.
PROTEIN - PROTEIN INTERACTIONS (2004-08)
This project focused on protein docking, especially of metalloproteins and transient complexes, and the use of experimental data as a source of constraints for docking simulations. It was funded by the Fundação para a Ciência e Tecnologia, MCTES.
PROTEIN (2000-03): Improving Constraint Programming to Predict Protein Structure
This project focused on the application of constraint programming to processing protein structural data (especially NMR). It was funded by the Fundação para a Ciência e Tecnologia, MCTES.