Identifying key motions in TrpRS and other proteins
(Carter; Hsu, Snoeyink)

Tryptophanyl-tRNA synthetase (TrpRS) catalyzes tryptophan activation by ATP and its subsequent aminoacylation to tRNA. The structures of TrpRS have been solved in both open and closed states. They provide evidence for major ligand-dependent domain motion both in solution and in crystals. It is of great interest to understand such motion and relate them to the function of the protein. Our first step is to determine which torsional angles contribute significantly to the final motion. Our current method consists of the following steps:

  1. Estimate the noise level and select a window size s.
  2. Compute the difference between all contiguous fragments of length s along the amino-acid chain.
  3. Threshold by hysteresis to locate large changes and break the chain into quasi-rigid sub-chains.
  4. Further break the sub-chains using bisection until each sub-chain is rigid according to a threshold.

We are testing our method on several proteins, in addition to TrpRS, that are interesting structurally and functionally.

Once the torsional degrees of freedom are located, we construct plausible motion pathways between different protein conformations. Molecular dynamics simulation and normal mode analysis are two main approaches in the literature for this purpose, but we believe that the probabilistic roadmap method for motion planning, developed in the robotics community, will be more useful. The goal of motion planning is to compute obstacle-avoiding motion for objects with many degrees of freedom. In our problem, the moving object is a protein with selected torsional angles. High-energy states of the protein are unfavorable and can be regarded as obstacles. A key insight from the roadmap method is that the connectivity of the space can be effectively captured in a graph whose nodes represent randomly sampled object configurations and edges represent obstacle-avoiding transitions. As we modify the roadmap method to reflect the characteristics and protein motion and devise a sampling scheme that takes into account the energy of different protein conformations, we hope to elucidate the motion pathway of TrpRS.