about the group
The research interests in our group are a
combination of topics that span the domains of biology, chemistry, data informatics, and high performance
computing. We develop and use theoretical methods, computational simulations along with data analysis techniques
to investigate biomolecules. We are particularly interested in biomolecules that have implications for clean
energy, environment and human health. We closely work with experimental collaborators for development and
validation of our computational models.
In the area of biology and chemistry, at the molecular level we are interested in allosteric
modulator design for development of safer medicines, and enzyme design and engineering for industrial
applications. Using a combination of computational techniques developed in the group over the last decade, we
have developed a biophysical model of how enzyme catalysis works. This model provides new insights into how the
solvent-enzyme coupling drives enzyme catalysis through long-range interactions in the enzyme structure. This
biophysical model is being applied (in collaboration with industrial partners) to development of new allosteric
modulators and hypercatalytic enzymes.
On the other side, in the area of computational science and engineering we are actively involved
in designing new computational algorithms and software performance optimization on emerging hardware
architectures (including heterogeneous architectures with FPGAs and GPUs), and open source software development.
We have optimized and ported molecular dynamics code on GPUs with 20 fold speed-up compared to
the CPU-only code. The insights gained are being applied to optimize popular codes and methods on the current
and future heterogeneous architectures. We are also investigating fault-tolerance and performance auto-tuning
strategies for improved end-user productivity on the future computing platforms.
Other software developed in the
group include VigyaanCD, an open source software workbench for bio/chemical modeling.