Computation and Machine Learning for Chemistry

Molecular simulations provide deep insight into chemical processes beyond what can be directly measured experimentally, holding major promise for accelerating the discovery of molecules and materials. In this collection we highlight a selection of recent computational studies published in Nature Communications, featuring advances in computational chemistry methods, applications to materials modelling, and progress towards machine-learning models that are transferable to new chemical processes and systems.

Molecular orbitals representing charge-transfer exciations delocalized over molecules in an organic molecular crystal.


Advances in computational chemistry

Computational chemistry for materials modelling

Machine learning development

Machine learning for chemical discovery