Idil Ismail, a current 3rd year HetSys student and her supervisor Scott Habershon have had their latest paper featured as an Editors' Pick in the Journal of Chemical Physics.

To date, several reaction discovery tools have been developed to accelerate and automate mechanistic elucidation. However, despite significant progress in this area, there is no single approach capable of rapidly assessing key kinetic parameters such as activation energies, without resorting to expensive first-principles calculations. In this paper, Idil and Scott show how machine learning (ML) can be used to accelerate chemical discovery pipelines; and evaluate the impact of the uncertainty associated with ML prediction of activation energies on the observable properties of chemical reaction networks based on microkinetic simulations using ML predictions.

You can read the article online here.

You can read more about Idil and Scott's work here.

We are delight to announce that three HetSys students were awarded prizes at the recent Midlands Computational Chemistry Meeting held in Manchester.

This meeting, organised by the Chilton Group in the Department of Chemistry at The University of Manchester is designed for PhD students and PDRAs to present talks and posters as well listen to keynote talks from notable academics in the field of Computational Chemistry.

Joe Gilkes (Cohort 2) was awarded first poster prize for: Stacking the odds: Distribution-biased generative deep learning for targeted design of organic electronics

Omar Adesida (Cohort 2) was awarded second poster prize for: Exploring the Phase Space of Hard Sphere Dimers Using Nested Sampling

And Idil Ismail (Cohort 1) was awarded second prize for her talk: High throughput screening of mechanistic hypotheses using machine learning and multi criteria decision making

On Monday 30th May, HetSys looks forward to welcoming Gus Hart (BYU) to give the Keynote Seminar: Building Useful Machine-Learned Interatomic Potentials

All are welcome, please sign up via the website.