Trent Barnard
Trent graduated in 2023.
He worked on using Machine Learning techniques to predict the stability of amorphous drugs with Dr Gabriele Sosso.
Amorphous drugs have many benefits when it comes to administering pharmaceutical drugs, the most important benefit is the increased bioavailability. Unfortunately, amorphous drugs are not widely used due to their tendency to recrystallise, if it were possible to accurately predict how long it would take these drugs to recrystallise then we could see a whole new wave of more effective pharmaceuticals.
Education
- 2019 - 2022: PhD in Mathematics for Real-World Systems
- 2018 - 2019: MSc in Mathematics for Real-World Systems
- 2015 - 2018: BSc in Mathematics from Coventry University
Publications
We published a paper that shows that a carefully selected group of descriptors often outperforms a large array of 'Standard Descriptors' in the context of machine learning for drug discovery.
(Under review) - This work is an attempt to understand the 'Boson Peak' of Tetrabutyl orthosilicate (TBOS). I performed the molecular dynamic simulations in this publication that were used to verify the physical experiments.
Research interests
My main field of interest is machine learning and more specifically all different flavours of neural networks. I have also become interested in optimisation methods and have recently published an open-source software python package for optimising chemical descriptors using genetic algorithms.
The code can be found at: https://github.com/gcsosso/SOAP_GAS