Adaptive QM/MM simulations

Increasing the size and complexity of atomistic systems in molecular simulations
is the driving force behind development of multi-scale techniques which
describe different regions of the system at different levels of
theory, e.g. using quantum mechanics (QM) where high accuracy is necessary and
Molecular Mechanics (MM) where such a lower level of description is adequate, but
taking advantage of the high speed of MM models is essential due to the overall size of the system.

Chemical processes such as reactant transport, diffusion,
membrane transport, ionic conductance, corrosive fracture of materials
or solvated chemical reactions, when modelled using such a QM/MM approach
all require reassignment of the QM and MM regions during the course of the
simulation. There is a unique opportunity to bring together three
participants, each of whom are in possession of an essential
component, to implement and widely disseminate a novel software tool.

The integration of adaptive QM/MM techniques
into the AMBER software will benefit a very large user base: with over
8,000 downloads from unique IP addresses of the AMBER Tools package
and >500 sites which use the AMBER MD engines it is clear that this
addition will have a wide reach. This project will therefore have an
extremely broad impact.

Researchers who are currently using AMBER to study biological and biochemical
systems using QM/MM will now be able to automate the selection of the QM region.
Further researchers who have not been using such techniques due to the lack
of adaptivity will now be able to.

Beyond academia, these users are primarily in the pharmaceutical and the wider
healthcare industries. These fields rely heavily on atomic scale modelling.