|
|
|
Search the Sherrill Group
|
|
|
|
Our group develops new approximations in electronic
structure theory, implements these approximations as
efficient computer programs (written in C++ and Python), and
applies these methods
to study challenging chemical problems. We are particularly
interested in new methods for electron correlation and
non-covalent interactions.
We are
studying the fundamental forces of molecular recognition
in prototype molecular systems
(including π-π, CH/π, SH/π, etc., interactions)
to determine their strength, geometric dependence, and
substituent effets. These questions are foundational for
rational drug design or crystal engineering.
We have developed the world's fastest wavefunction-based symmetry
adapted perturbation theory (SAPT) program and are using it to
understand the physical nature of non-covalent interactions from
small systems like benzene dimer to large systems like proflavine/DNA
intercalation complexes. One key finding from our recent studies is
that π-π interactions in DNA feature very significant
contributions from charge-overlap ("charge penetration")
terms that are not captured in classical force field models.
We specialize in developing modular, open-source software, and
methods to automatically generate, run, and analyze large numbers
of computations for analyzing approximate methods, for data-driven
approaches, and for machine-learning applications.
See our publication list for
more details.
We are one of the primary
developers of the Psi
program package.
The Sherrill group has significant computing resources
through a local, group-built cluster, and also through
the Georgia Tech
Hive supercomputer, for which Dr. Sherrill
is one of the co-PI's.
© The Sherrill Group
Georgia Institute of Technology
Main