seed projects

Sapna Sarupria: Predictive Computation to Design Novel Peptides for Hydrogel Formation

Using machine learning to enhance simulations of self-assembling molecules" Description: Self-assembly of polymers and peptides can be engineered by changing the assembling molecules. This can be exploited to get a broad range of structures and properties. However, predicting the self-assembled structure a priori based simply on the molecular properties is rather challenging. Molecular simulations can be powerful tools to make such predictions but simulations are challenged by two aspects: the large design space cannot be easily traversed and often the length and timescales involved in self-assembly make the computations expensive. In our work, we are focused on combining machine learning and molecular simulations to develop novel tools that will address both these challenges and thus enable the discovery of peptides/polymers for desired self-assembly behavior.

Funded by the National Science Foundation through the University of Minnesota MRSEC under Award Number DMR-2011401