Dr Fergus Boyles

Research Software Engineer

About Me

I started my academic adventure with mathematical physics at the University of Edinburgh, eventually focusing on statistical physics and mathematical modelling of biological systems. Afterwards, I hit the snooze button on the alarm clock of life and did my DPhil in Systems Biology here in Oxford, in which I focused on applying machine learning to protein-ligand interactions. Now I'm a research software engineer in the Oxford Protein Informatics Group, where I focus on developing research projects into robust software that can be deployed to production both in our group and to the wider world through web applications and commercially-licenced software packages.

Research Interests

My own research interests lie in the realm of small molecule drug discovery, in particular predicting protein-ligand binding. In practice, most of my time is spent developing immunoinformatics tools and web applications in collaboration with the rest of the Oxford Protein Informatics Group.

Publications

Abanades Kenyon, B., Wong, W., Boyles, F., Georges, G., Bujotzek, A. and Deane, C. (2023) “ImmuneBuilder: Deep-Learning models for predicting the structures of immune proteins”, Communications Biology, 6.

Durant, G., Boyles, F., Birchall, K., Marsden, B. and Deane, C. (2023) “Robustly interrogating machine learning-based scoring functions: what are they learning?”, bioRxiv.

Moesser, M., Klein, D., Boyles, F., Deane, C., Baxter, A. and Morris, G. (2022) “Protein-ligand interaction graphs: Learning from ligand-shaped 3D interaction graphs to improve binding affinity prediction”, bioRxiv.

Abanades, B., Wong, W., Boyles, F., Georges, G., Bujotzek, A. and Deane, C. (2022) “ImmuneBuilder: Deep-Learning models for predicting the structures of immune proteins”, bioRxiv.

Olsen, T., Boyles, F. and Deane, C. (2021) “Observed Antibody Space: A diverse database of cleaned, annotated, and translated unpaired and paired antibody sequences”, Protein Science, 31(1), pp. 141–146.

Boyles, F., Deane, C. and Morris, G. (2021) “Learning from Docked Ligands: Ligand-Based Features Rescue Structure-Based Scoring Functions When Trained on Docked Poses”, Journal of Chemical Information and Modeling [Preprint].

Boyles, F., Deane, C. and Morris, G. (2021) “Learning from docked ligands: ligand-based features rescue structure-based scoring functions when trained on docked poses”, Journal of Chemical Information and Modeling, 62(22), pp. 5329–5341.

Boyles, F., Deane, C. and Morris, G. (2019) “Learning from the ligand: using ligand-based features to improve binding affinity prediction”, Bioinformatics, 36(3), pp. 758–764.

Boyles, F., Deane, C. and Morris, G. (2019) “Learning from the ligand: using ligand-based features to improve binding affinity prediction.”

Boyles, F., Deane, C. and Morris, G. (2019) “Learning from the Ligand: Using Ligand-Based Features to Improve Binding Affinity Prediction.”

Contact Details

Email: fergus.boyles@stats.ox.ac.uk

Office: 2.03

Research Groups

Oxford Protein Informatics Group

Research Students