Abstrakt

One of the most important challenges in performing Molecular Dynamics (MD) simulations of large protein complexes is to accommodate the model accuracy and the simulation timescale. Hitherto, for the most relevant dynamics of protein aggregates in an explicit aqueous environment, the timescale reachable for the all-atoms simulations is of hundreds of nanoseconds. This range is four to six orders of magnitude smaller than processes occurring within the cell and thus computational results cannot be directly compared to experimental results. In order to increase the simulation performance, a less detailed description of the system could be used but it should be considered that the more simplified model used, the less accurate and predictive the results are. Nevertheless, as the fast and slow molecular motions are quite independent, it is possible to ignore fast vibrations for the study of slow motions. Thus, a simplified description can be applied to analyze the mechanisms, dynamics and structural changes of a protein.The principle of coarse-grained (CG) models is to reduce the number of degrees of freedom of the system by integrating some of them into a smaller number. A molecule is then described by the interaction sites representing groups of atoms - beads. Very different levels of CG models can be applied, ranging from the "united atoms" model, where only non-polar hydrogen atoms are ignored, to mesoscale approaches where the molecule is reduced to few rigid regions of well-defined equilibrium structures identified as the natural coarse grained elements. Therefore, applying CG models can significantly reduce the resolution description of the system and, consequently, increase the computations efficiency - the smaller the beads number is, the less computationally expensive the simulation and, consequently, the larger timescale can be applied.The method of coarse-graining has been known for years. The main challenge and limitation of this technique has always been to properly parameterize the force field. On one hand it should be accurate: effectively describing specific interactions with few parameters and functional forms; on the other hand, some extent of versatility is required: the force field must be independent from the reference configuration and transferable to different systems. Recently, some main advances in force field parameterization have been made, making it possible to achieve the order of accuracy comparable to the results obtained from atomistic models. The CG method, reinforced by enhanced computer power have recently become a point of high interest for the large, solvated protein dynamics study.

Autorzy (3)