Coarse-grained Computational Modeling of Actin Polymerization
Summary
A systematic understanding of the biochemical and biophysical mechanisms that govern actin organization is important for both medical applications as well as the development and design of advanced biomimetic self-organizing materials. Formins are importants proteins for cells with the ability to remain processively associatied with the actin filament end, adding profilin-actin subunits to the actin filament. The aim of this project is to develop computational models of how formins accelerate actin polymerization at the molecular level.
Job Description
The student will perform molecular dynamics simulations with LAMMPS, using a coarse-grained model in which each aminoacid is replaced by a bead, interacting with other beads through the Kim and Hummer potential. Folded segments of actin, profilin and formin will be modeled as rigid bodies. Serial molecular dynamics and enhanced sampling metadynamics (PWTE) simulations will be employed.
Computational Resources
Typical simulations with actin binding/unbinding will involve 20 independent simulations each of 10 microseconds of simulated time. Each simulation would involve usage of a few nodes in Comet. The modeled system is large enough to benefit from parallelization, but small enough that the jobs do not need no more than a few nodes. Both the shared and compute nodes will be used. The shared queue will be used for submission of large numbers of small jobs, and the compute queue will be used for simultaneously running multiple, independent parallel jobs.
Contribution to Community
Position Type
Apprentice
Training Plan
The student will be mentored by the PI (Dimitrios Vavylonis) and graduate student Brandon Horan, who has experience using the proposed methods with one published paper, in collaboration with Prof. Jeetain Mittal (Lehigh Chemical Engineering, XSEDE member). The student will be trained in LAMMPS and usage of XSEDE. The student will participate in weekly group meetings while a google document will be used for continuous monitoring of the project. A working mentoring and research advising has already been established with the proposed student who is majoring in Physics and Computer Science. The student has office space and a dedicated Linux system in the Vavylonis group. He has been trained in the use of LAMMPS and XSEDE since the summer of 2018. He has many results that he plans to present at the Annual meeting of the Biophysical Society in 2019 and as co-author of a publication.