Our main research directions overlap. We welcome related questions that connect methods, data and meaningful applications.
Predictive modelling, data curation and evaluation methods for applied research. We are interested in how data quality, study design and validation affect the conclusions a model can support.
Research datasets, reproducible analysis and the evaluation of predictive models.
Graph neural networks, transformers, multimodal learning and generative models. We study representations and learning strategies for complex scientific data.
Representation learning, complementary data sources and comparisons with strong baselines.
Computational analysis of biological sequences and molecular data, including peptide prediction and the identification of functional sequence features.
Biological datasets, sequence-based prediction and computational questions supported by experimental context.
Data science and AI applied to health information and clinical questions. Our interests include medical prediction, human–computer interaction and the careful evaluation of technologies for healthcare.
Clinically grounded questions, external validation and evidence synthesis.
Molecular property prediction, drug–target modelling and computational approaches to molecular design. Our published work also includes molecular data resources and curation software.
Molecular datasets, computational screening and modelling linked to experimental follow-up.
Let’s connect
Students, researchers and collaborators are welcome to get in touch.
