Wires Within Wires A Minimal Model for Computational Bioelectronic Peptide Design R. A. Mansbach 1 A. L. Ferguson 2 1 Physics Department 2 Materials Science Department University of Illinois at Urbana-Champaign Blue Waters Symposium, Sunriver, OR, June 4, 2018
π -conjugated self-assembling optoelectronic peptides Wires Within Wires Mansbach, Rachael Motivation Patchy Model Results Conclusions and Future Work Wall, Brian D., et al. “Supramolecular Polymorphism: Galagan, Y.,& Andriessen, R. (2012). Tunable Electronic Interactions within π -Conjugated “Organic photovoltaics: technologies and Peptide Nanostructures Dictated by Primary Amino Acid manufacturing.” INTECH Open Access Publisher. Sequence.” Langmuir30.20 (2014): 5946-5956. 2 / 15 topic-apple-watch-all.png?itok=OUtlCphV
π -conjugated self-assembling optoelectronic peptides Wires Within Wires Mansbach, Rachael Motivation Patchy Model Results Conclusions and Future Work Wall, Brian D., et al. “Supramolecular Polymorphism: Galagan, Y.,& Andriessen, R. (2012). Tunable Electronic Interactions within π -Conjugated “Organic photovoltaics: technologies and Peptide Nanostructures Dictated by Primary Amino Acid manufacturing.” INTECH Open Access Publisher. Sequence.” Langmuir30.20 (2014): 5946-5956. topic-apple-watch-all.png?itok=OUtlCphV 2 / 15
π -conjugated self-assembling optoelectronic peptides Wires Within Wires Mansbach, Rachael Motivation Patchy Model Results Conclusions and Future Work Wall, Brian D., et al. “Supramolecular Polymorphism: Galagan, Y.,& Andriessen, R. (2012). Tunable Electronic Interactions within π -Conjugated “Organic photovoltaics: technologies and Peptide Nanostructures Dictated by Primary Amino Acid manufacturing.” INTECH Open Access Publisher. Sequence.” Langmuir30.20 (2014): 5946-5956. 2 / 15 topic-apple-watch-all.png?itok=OUtlCphV
DXXX series demonstrates hierarchical assembly Wires Within Wires Optical Clusters Mansbach, Rachael Motivation Patchy Model Results Conclusions and Future Work 3 / 15
DXXX series demonstrates hierarchical assembly Wires Within Wires Contact Clusters Optical Clusters Mansbach, Rachael Motivation Patchy Model Results Conclusions and Future Work 3 / 15
Reaching longer time and length scales Wires Within Wires Mansbach, Rachael Motivation Minimal coarse-grained model Patchy Model Large computational infrastructure Results Do parameter sweep over well Conclusions and Future depths and radii to gain Work understanding of effect of different interaction parameters on assembly at mesoscale 4 / 15
Reaching longer time and length scales Wires Within Wires Mansbach, Rachael Motivation Minimal coarse-grained model Patchy Model Large computational infrastructure Results Do parameter sweep over well Conclusions and Future depths and radii to gain Work understanding of effect of different interaction parameters on assembly at mesoscale 4 / 15
Understanding chemical interactions at low resolution Wires Within Wires Mansbach, Rachael Motivation Minimal coarse-grained model Patchy Model Results Large computational infrastructure Conclusions and Future Do parameter sweep over well Work depths and radii to gain understanding of effect of different interaction parameters on assembly at mesoscale 5 / 15
Aggregate shape and fractal dimension match previous work Wires Within Wires Mansbach, Rachael Motivation Patchy Model Results Conclusions and Future Work Ardona, Herdeline Ann M., and John D. Tovar. “Energy transfer within responsive π -conjugated coassembled peptide-based nanostructures in aqueous environments.” Chemical Science 6.2 (2015): 1474-1484. 6 / 15
Interaction parameters control aggregate morphology Wires Within Wires Mansbach, Rachael Motivation Patchy Model Results Conclusions and Future Work Increasing side chain stickiness increases disorder Side chain size controls transition between flat ribbon and twisted fibril 7 / 15
Optical cluster growth is primarily controlled by side chain interactivity Wires Within Optical Cluster Growth Wires Increasing side Mansbach, Rachael chain well depth Motivation increases Patchy Model favorability of side Results chain–side chain Conclusions interactions and Future Work Biggest increase as side chain interactivity decreases below core–core interactivity 8 / 15
Side chain radius affects contact cluster growth more strongly Wires Within Contact Cluster Growth Fewer configurations Wires Mansbach, Rachael Motivation Patchy Model Results Conclusions and Future Work Increasing cross-section 9 / 15
Identification of optimal parameter sets Wires Within Wires Mansbach, Rachael Motivation Patchy Model Results Conclusions and Future Work Pareto frontier Tradeoff between different objectives 10 / 15
Five candidates flagged for future study Wires Within Wires Mansbach, Rachael Motivation Patchy Model Results Conclusions and Future Work 11 / 15
Next steps: Active Learning Wires Within Wires Mansbach, Rachael Motivation Patchy Model Results Conclusions and Future Work Brochu, Eric, Vlad M. Cora, and Nando De Freitas. “A tutorial on Bayesian optimization of expensive cost functions, with application to active user modeling and hierarchical reinforcement learning.” arXiv preprint arXiv:1012.2599 (2010). 12 / 15
Next steps: Active Learning Wires Within Wires Mansbach, Rachael Motivation Patchy Model Results Conclusions and Future Work Brochu, Eric, Vlad M. Cora, and Nando De Freitas. “A tutorial on Bayesian optimization of expensive cost functions, with application to active user modeling and hierarchical reinforcement learning.” arXiv preprint arXiv:1012.2599 (2010). 12 / 15
Next steps: Active Learning Wires Within Wires Mansbach, Rachael Motivation Patchy Model Results Conclusions and Future Work Brochu, Eric, Vlad M. Cora, and Nando De Freitas. “A tutorial on Bayesian optimization of expensive cost functions, with application to active user modeling and hierarchical reinforcement learning.” arXiv preprint arXiv:1012.2599 (2010). 12 / 15
Why Blue Waters? Wires Within Wires Mansbach, Rachael Scale of problem Motivation 300 simulations of 10,648 monomers Patchy Model Each simulation requires multiple Results GPU acceleration and produces Conclusions and Future 10-20 gigabytes of data to be Work analyzed Big data research infrastructure Access to broader big data community https://www.slideshare.net/sergejsgroskovs/ pragmatism-philosophy-of-science-lecture-slides 13 / 15
Why Blue Waters? Wires Within Wires Mansbach, Scale of problem Rachael 300 simulations of 10,648 Motivation monomers Patchy Model Each simulation requires multiple Results GPU acceleration and produces Conclusions 10-20 gigabytes of data to be and Future Work analyzed Big data research infrastructure Access to broader big data https://www.slideshare.net/sergejsgroskovs/ community pragmatism-philosophy-of-science-lecture-slides 13 / 15
Broader Impact Wires Within Wires Mansbach, Rachael Created a patchy model that Motivation recapitulates DXXX properties and Patchy Model reaches mesoscopic scale Results Showed effects of changing Conclusions and Future parameter space Work Identified potential ways to design for optimal parameters Part of a multiscale model for rational peptide design 14 / 15
Acknowledgments Wires Within Wires Mansbach, Rachael Motivation Patchy Model Results Conclusions and Future Work *This research is part of the Blue Waters ∗ sustained-petascale computing project, which is supported by the National Science Foundation (awards OCI-0725070 and ACI-1238993) and the state of Illinois. Blue Waters is a joint effort of the University of Illinois at Urbana-Champaign and its National Center for Supercomputing Applications. 15 / 15
Backup Slides Wires Within Wires Mansbach, Rachael Choice of parameter space for sweep Examples of single-parameter computations Additional data Ideal Gas Model of Aggregation 1 / 13
Initial Parameter Sweep: Aromatic Cores Wires Within Non cofacial aromatic ǫ BB Cofacial aromatic ǫ A Wires Mansbach, Rachael Choice of parameter space for sweep Examples of single-parameter computations ~ 18 Additional data Ideal Gas Model kT of Aggregation Cv Set to 1 k B T Sweep over 2.5-7.5 k B T depth 2 / 13
Initial Parameter Sweep: Side Chains Wires Within Side chain ǫ SC Wires Side chain σ SC Mansbach, Rachael Choice of parameter space for sweep Examples of single-parameter computations ~ 2 kT Additional data Ideal Gas Model of Aggregation Sweep over 1.0 -1.75 nm Sweep over 0.2-10 k B T 3 / 13
Example of growth rate calculations Wires Within Wires Mansbach, Rachael Choice of parameter space for sweep Examples of ǫ A = 2 . 5 k B T single-parameter computations Additional data σ SC = 1 . 5 nm Ideal Gas Model of Aggregation Main Text Backups 4 / 13
Recommend
More recommend
