A major focus of biology over the last century has been identifying the
molecules of life. In the last five years, significant progress towards
cataloging this molecular library has been achieved. The next challenge
for biomedical research is to understand how these biomolecules interact
with one another. Because of the staggering number of potential interactions,
a combination of experiment and computation is the only practical approach
to the identification and characterization of molecular complexes and networks.
Our broad goals are to dramatically improve access to protein structure
modeling and docking, amplify structural reasoning in biology, integrate
experiment and computation, provide new viewpoints derived from large-scale
interaction maps, and facilitate health-related research.
Specifically, we are interested in a comprehensive description of the interactions
between proteins and their ligands. Towards this aim, we propose to construct
a widely accessible software and hardware system for genome-wide mapping
of the interactions of proteins with small molecules ( ie , ligand
docking) and with other proteins ( ie , protein-protein docking).
This system will consist of two specialized computational pipelines, one
for small molecule ligands and another for protein-protein interactions.
The input to the small molecule docking
pipeline will be a set of protein sequences, a database of experimental
protein structures, and databases of small molecules; the output
will be the prediction of likely ligands for the proteins and proposed
3D structures for their complexes.
The input to the protein-protein docking pipeline will be a set of protein
sequences, a database of experimentally determined protein structures, and
a set of known protein-protein interactions; the output will be low- and
high-resolution models of protein complexes as well as estimated specificities,
depending on the information available about the interacting proteins and
their complexes.
To construct the pipeline, we have assembled an outstanding team of researchers
who are well-versed in the required technology and are recognized experts
in their fields. Their laboratories have produced pioneering software packages
that have been applied to important scientific problems. Frequently, these
packages have been distributed to other academic and industrial laboratories.
In addition, we have noted computer scientists, young investigators, and
a strong selection of experimentalists joining in frontline biological collaborations
that help motivate and create the pipeline.
