|
1200 E. California Blvd.
Pasadena, CA
91125-9600
Mail Code: 114-96
Location: 151 Broad
Phone: (626) 395-8084
|
Research
With our current design software, we can score the stabilities of various protein
configurations very quickly and use this information to select proteins with
desired properties. The scoring functions used with protein design algorithms
rely on the use of rotamer libraries for amino acid sidechain placements and
treat the effects of solvent implicitly, usually by evaluating surface areas
of solvent-exposed atoms. At the end of the initial design process, we generate
a number of viable 3D protein configurations (on the order of tens to hundreds)
that are candidates for synthesis and experimental validation. We are now developing
a follow-on procedure to reevaluate these final candidate configurations with
more extensive scoring functions that treat the effects of the solvent explicitly
and also relax the requirement that the sidechain positions conform to a given
set of rotamers. The computational requirements for this new procedure are substantial,
but the expectation is that by using a more rigorous scoring function we can
improve the probability that we synthesize and validate the more useful configurations
from the candidates generated by the design algorithms.
With the new methods, we explicitly add solvent molecules to completely envelope
the candidate protein configurations. The scoring function estimates the relative
stabilities of various protein configurations, which depends upon their internal
structure and their additional stabilization or destabilization due to solvent
interactions. Because we have explicit solvent atoms and their positions in the
calculation, we are able to calculate the interaction between protein and solvent
directly. In addition, we estimate the impact the presence of the protein has
on causing the solvent water molecules to reorganize and add this contribution
to the overall protein configuration score. It is essential that the combined
system, protein and solvent, be allowed complete conformational freedom to allow
the explicit water molecules to pack tightly to the protein surface, and for
the amino acid sidechains to optimize their positions. Results to date indicate
that these three general components (internal energies, protein interactions
with solvent, and solvent reorganization) all contribute comparably for folded,
stable proteins.
home | research | lab
members | publications | contact | links | group
only
© Calfornia Institute of Technology
|
