Lab Home | Phone | Search
Center for Nonlinear Studies  Center for Nonlinear Studies
 Home 
 People 
 Current 
 Executive Committee 
 Postdocs 
 Visitors 
 Students 
 Research 
 Publications 
 Conferences 
 Workshops 
 Sponsorship 
 Talks 
 Seminars 
 Postdoc Seminars Archive 
 Quantum Lunch 
 Quantum Lunch Archive 
 P/T Colloquia 
 Archive 
 Ulam Scholar 
 
 Postdoc Nominations 
 Student Requests 
 Student Program 
 Visitor Requests 
 Description 
 Past Visitors 
 Services 
 General 
 
 History of CNLS 
 
 Maps, Directions 
 CNLS Office 
 T-Division 
 LANL 
 
Tuesday, January 16, 2018
09:00 AM - 10:00 AM
CNLS Conference Room (TA-3, Bldg 1690)

Seminar

From Atoms to Aggregates: Machine Learning and Coarse-graining for Simple Models of Complex Biophysical Phenomena

Rachael A. Mansbach
University of Illinois at Urbana-Champaign

Protein folding and aggregation are highly complex biophysical phenomena that can naïvely be described in the 3N-dimensional space of the Cartesian coordinates of the atoms constitut- ing the system. Such a representation, however, is computationally expensive and reveals little physical insight. In order to understand the effects of external factors such as chemistry and environment on folding and assembly of peptides, it is necessary to employ or derive simplified descriptions of the problems. In this talk, I focus on two problems: (i) the effects of chemistry the configurations of a set of peptides with antimicrobial properties in aqueous solution and (ii) the effects of flow, pH, and chemistry on the self-assembly of a set of peptides with opto- electronic properties. I modify the machine learning technique known as the diffusion map for dimensionality reduction to compare the configurations of antimicrobial peptides with differ- ent side chain lengths and reveal a critical side chain length controlling whether the backbone takes on a helical or random coil configuration in water. I employ bottom-up and top-down coarse-graining to reach the time and length scales necessary to observe mesoscopic assembly of a set of optoelectronic peptides and characterize the effects of pH, flow, and chemistry on their growth kinetics and morphology. I demonstrate that assembly proceeds hierarchically, that flow has little effect up to time scales of hundreds of nanoseconds, and that controlling the pH controls the orderedness of aggregation. I show that less interactive side chains lead to more linear, ordered aggregates. Overall, the work herein demonstrates several new techniques for understanding the effects of chemistry and environment on complex biophysical systems, leads to greater understanding of the configurations adopted by antimicrobial peptides, and is a strong step in the direction of rational design of self-assembling peptides for bioelectronic applications such as organic photovoltaic cells, organic field effect transistors, and biocompatible pH sensors.

Host: Gnanakaran, Sandrasegaram