Postdoctoral Fellow
Research group | The Medicinal Chemistry Group
Main supervisor | Osman Gani
Co-supervisor |?-
Affiliation | Department of Pharmacy, UiO?
Contact | neil.thomson@farmasi.uio.no
Short bio
My academic journey bridges the physical and life sciences: I completed my BSc in Physics and Philosophy at the University of Aberdeen and my MSc in Theoretical Physics at the University of Glasgow, before shifting toward biology with my PhD in Computational Biology at the University of Dundee. During my doctorate, I gained industry experience at Pharmacelera, where I contributed to developing quantum mechanics-based software for drug screening applications. I am currently a postdoctoral research fellow at the University of Oslo under the DSTrain MSCA Co-fund Programme, working with the Gani Lab in the Department of Pharmacy on AI-integrated computational drug discovery.
Research interests and hobbies
My research sits at the intersection of physics-based molecular modeling and machine learning for drug discovery. I develop and apply computational pipelines that combine molecular dynamics simulations, free energy calculations, and AI/ML to understand protein-ligand interactions and accelerate the identification of promising drug candidates. A particular focus is G protein-coupled receptors (GPCRs) and understanding the structural determinants of signaling bias - how different drugs acting on the same receptor can produce distinct physiological outcomes. Outside of research, I enjoy staying fit, spending time with friends and being in nature.
DSTrain project
LOCKSMITH, A Computational Tool for Predicting Biased Signalling in GPCRs
Many drugs that target G protein-coupled receptors (GPCRs) cause unwanted side effects because they cannot selectively engage a single intracellular signalling pathway. Biased agonists - ligands that preferentially activate one pathway over another - offer a solution, but identifying them experimentally is costly, time-consuming, and resource-intensive. LOCKSMITH aims to develop an in silico cell signalling assay capable of predicting bias factors for GPCR ligands, streamlining the discovery of functionally selective compounds. The project combines molecular dynamics simulations, AI and machine learning to identify conformational signatures of biased signalling and infer bias levels entirely computationally. By building functional selectivity directly into the virtual screening process, LOCKSMITH has the potential to accelerate the discovery of safer, more targeted therapeutics.
?
Publications
DSTrain publications
Please see https://orcid.org/0000-0002-7062-2716
?
Previous publications
?
?