Investigating the repulsive behaviour of topological atoms under compression

The Quantum Chemical Topology group is using machine learning to develop a computer model (based in Fortran) that can accurately simulate the behaviour of proteins, specifically with the aim of more fundamentally understanding the onset of Alzheimer’s disease.

My research looked at the energetics of the compression of topological atoms and involved testing whether the results were consistent with more traditional models of atoms.

I was involved the design and execution of a set of computational experiments which often had to be created in Python from scratch. These were submitted to Gaussian using bash before being run and analysed on a high-performance computer cluster. Data analysis was done both locally using Numpy and Matplotlib and on a distributed computing cluster using bash.

See the paper here.

Front cover featuring the paper in May 2019