Frontiers of Coarse-graining

Gregory Voth of the University of Chicago is supported by an award from the Chemical Theory, Models and Computational Methods program in the Division of Chemistry to develop and apply systematic theory and computer simulation methods that can bridge the various length and time scales inherent in complex systems, in order to model their overall behavior. The objective of these new multi-scale theory and simulation methods is to connect molecular behaviors at small length and time scale scales with the phenomena they cause at significantly larger length and time scales, in systems such as in biological cells, new materials, and liquids.

The broader impacts of the project include STEM (Science, Technology, Engineering and Mathematics) development of members of underrepresented groups through active mentorship. In the longer term, this research also has the potential to improve the well-being of society (i) by helping to lay the fundamental groundwork for computer simulation-aided advances in the health sciences (e.g. modeling the molecular basis of viral infection) and in materials science (e.g. new biologically-inspired materials),(ii) by building increased partnerships between academia and industry (e.g. to model therapeutic antibody formulations via simulation models), and (iii) by providing enhanced infrastructure to the community in providing the basis for open source software devoted to multi-scale computer modeling.

This research project from the Voth group at the University of Chicago is focused upon the development of theoretical and computational methodology to study complex condensed phase systems, including biomolecules, soft materials, and liquids, in new ways. The conceptual basis for the research relates to expanding the frontiers of systematic coarse-graining methods at various levels of resolution, as well as expanding their application to key macromolecular problems.

The mathematical basis for the project lies in the implementation of equilibrium and non-equilibrium statistical mechanics and dynamics for multi-scale modeling. The project will also develop transformative computer simulation approaches from these theoretical advances to help reveal and understand the molecular-level features that define the functional behavior of, e.g., biomolecular systems at large length and time scales.

Various connected resolutions of the theoretical framework will be considered, including the highly (ultra) coarse-grained as well as the more finely resolved coarse-grained. New techniques from machine learning will be introduced to develop models and quantify errors, as well as new computer algorithms implemented to define and solve coarse-grained molecular models.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.