Seminar: Graph Theoretical Descriptors for Biomimetic Nanoparticles and Fibrous Nanocomposites

Zoom ID: https://cuboulder.zoom.us/j/96007553384

Password: i-aim

Nicholas Kotov (U Michigan)
“Graph Theoretical Descriptors for Biomimetic Nanoparticles and Fibrous Nanocomposites”

Abstract: Descriptors based on graph theory (GT) are needed to achieve accurate representations of two classes of nanostructures for the successful application of machine learning (ML). First, a method to depict protein structure at molecular, nanoscale, and sub-microscale levels is described to predict complex formation and organization of protein-nanoparticle interfaces using several ML-algorithms. Second, a methodology to utilize GT descriptors in nanofibrous composites is developed. The computational package Structural GT is introduced to automatically produce a GT description and structural descriptors of percolating nanoscale networks from micrographs.

About the Speaker: Nicholas Kotov is the Irving Langmuir Distinguished Professor of Chemical Sciences and Engineering at the University of Michigan. He demonstrated that the ability to self-organize into complex structures is the unifying property of all inorganic nanostructures. He developed a family of bioinspired composite materials with a wide spectrum of properties that were previously unattainable in classical materials, such as nacre-like ultrastrong, transparent composites, enamel-like, stiff yet vibration-isolating composites, and cartilage-like membranes with high strength and ion conductance.