Self-assembling peptides hold promise as versatile, environmentally benign nanoscale materials and scaffolds, but it remains challenging to predict sequence-specific behavior without accurate coarse-grained models.  Are there theoretically intuitive and numerically robust strategies for turning all-atom simulations into coarse models that preserve sequence-specific physics? How can we identify what atomic details are unnecessary?  We discuss a new approach to this problem based on information theory.  A quantity called the relative entropy measures the information lost upon coarse graining and hence the (inverse) fitness of a coarse-grained model.  Minimization of the relative entropy provides a universal principle for coarse-graining, and a way to �automatically� discover coarse models of many systems.  This approach enables us to develop simple but surprisingly accurate peptide models, and permits large-scale simulations of several kinds of self-assembly: amyloid-like oligomerization and aggregation, interactions of tethered peptides on surfaces, and the formation of nanotubes by diphenylalanine peptides.

Understanding peptide assembly with coarse-grained models designed by information theory

245th National Meeting (2013)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.