Dr. Li-Tien Cheng, Department of Mathematics; University of California, San Diego
Abstract: The problem of the shape of a biomolecule involves studying a biomolecule, comprised of solute atoms, in its natural environment of a solvent that is roughly salt water. Of interest is where the water molecules are located surrounding the biomolecule. This especially affects operations such as protein docking, which is of importance in, for example, tumor suppression. In the implicit solvation framework, the water is represented continuously, thus transforming the problem into one of finding the interface separating the biomolecule from the water, known as the solute-solvent interface. The desired location of this interface minimizes free energy, and a level-set method can be applied to this variational procedure to capture it. Such methods have been successful in other interface problems in, for example, fluid flow, materials science, image processing, and tumor growth. In implicit solvation, it can produce results comparable to those from molecular dynamics simulations, but with much faster simulation speeds, from several days down to a day. Here, we improve on this by introducing, instead, the binary level-set method, which can further speed up the simulations, from a day down to seconds, by removing the slow partial differential equation solves at the expense of accuracy. This allows us to study the protein docking process in more detail using Monte Carlo simulations, where tens of thousands of solute-solvent interfaces are generated at each step.
Bio: Dr. Li-Tien Cheng is a Professor of Mathematics at UCSD. He works in the area of scientific computation with an emphasis on computational partial differential equations and interface dynamics via the level-set method. He received his Ph.D. from the Department of Mathematics at UCLA in 2000.
Please contact the organizer for the Zoom link.