Richard Cooper

Increasing the performance, trustworthiness and practical value of machine learning models: a case study predicting hydrogen bond network dimensionalities from molecular diagrams

Mar 122020
 

CrystEngComm, 2020, DOI: 10.1039/D0CE00111B

The performance of a model is dependent on the quality and information content of the data used to build it. By applying machine learning approaches to a standard chemical dataset, we developed a 4-class classification algorithm that is able to predict the hydrogen bond network dimensionality that a molecule would adopt in its crystal form with an accuracy of 59% (in comparison to a 25% random threshold), exclusively from two and lower dimensional molecular descriptors. Although better than random, the performance level achieved by the model did not meet the standards for its reliable application. The practical value of our model was improved by wrapping the model around a confidence tool that increases model robustness, quantifies prediction trust, and allows one to operate a classifier virtually up to any accuracy level. Using this tool, the performance of the model could be improved up to 73% or 89% with the compromise that only 34% and 8% of the total set of test examples could be predicted. We anticipate that the ability to adjust the performance of reliable 2D based models to the requirements of its different applications may increase their practical value, making them suitable to tasks that range from initial virtual library filtering to profile specific compound identification.

A hexagonal planar transition-metal complex

Oct 172019
 

Nature 574, 390–393 (2019). [ doi: 10.1038/s41586-019-1616-2 ]

We report the isolation and structural characterization of a simple coordination complex in which six ligands form bonds with a central transition metal in a hexagonal planar arrangement. The structure contains a central palladium atom surrounded by three hydride and three magnesium-based ligands. This finding has the potential to introduce additional design principles for transition-metal complexes, with implications for several scientific fields.

  • Publisher’s copy: Nature 574, 390–393 (2019)

Noah Francis

Oct 042019
 

Noah is researching development of better descriptors of molecules for use in machine learning to prediction of crystal properties. Despite being stuck in the basement, Noah’s favourite building on the Chemistry estate is the CRL. His favourite functional is B3LYP and when not in the lab he plays hockey and 5-a-side football.

Aditya Desai

Oct 042019
 

Aditya is developing machine learning methods for predicting the classification of various crystallographic properties. He mostly works with Python 3 and associated ML and DL libraries.

In his spare time he enjoys baking and hiking. His go to space group is Fmmm and his favourite intermolecular interaction is pi-pi stacking – classic.

Natalie Haarer

Oct 042019
 

Natalie is implementing a scattering model for chemical groups which behave as “hindered rotors” to the base Fortran code of the CRYSTALS software. The representation requires fewer parameters and is more physically realistic than current models. Look out for better trifluoromethyl groups in future!

Natalie does not have a favourite programming language, believing instead that we should use the appropriate tool for the problem being solved. In her spare time she is a keen cross-country runner and has completed over 170 parkruns.

Samah Lakha

Oct 032019
 

Samah is continuing a project investigating accurate proton position determination in acid-base co-crystals.

In her spare time Samah is on the commitee for the 2019 LMH ball. Good luck!

 Older Entries  Newer Entries