Alexander Pak, Ph.D., Assistant Professor, Chemical and Biological Engineering, Colorado School of Mines

Speaker: 

Alexander Pak, Ph.D.

Department of Chemical and Biological Engineering

Quantitative Biosciences and Engineering Program, Materials Science Program

Colorado School of Mines

Faculty Host: Carrie Mills

Title: Data-Driven Molecular Dynamics of Self-Organizing Biomolecular Complexes

Abstract: 

Elucidating the relationships between microscopic interactions and macroscopic
behavior of self-assembled soft and biological matter has led to numerous advances
across energy, sustainability, and healthcare. Many such systems exhibit hierarchical
structures undergoing morphological transitions that often depend on the competition
between thermodynamic and kinetic factors. However, it remains largely unknown how
collective molecular reorganization may be modulated, which, in turn, may regulate
macroscopic functionality. In this context, I argue that the common biological (and
materials science) paradigm of sequence-(structure)-function relationships is incomplete
without a holistic view on dynamics. Throughout this talk, I will explore this theme by
leveraging our computational approach that combines atomistic and coarse-grained
modeling, the latter of which emulates the behavior of these systems under reduced
representations (i.e., simplified degrees of freedom), with data-driven approaches. First,
I will discuss two-dimensional para-crystalline lattices composed of surface-layer proteins where controlled disassembly is of interest as an anti-virulence strategy. Using a combination of molecular dynamics simulations, machine learning, and explainable artificial intelligence methods, I will reveal mechanisms that therapeutics can target to
promote surface-layer depolymerization and aid in the fight against antibiotic resistant bacteria. Next, I will share our recent work on the assembly of bacterial microcompartment shell proteins into diverse, polymorphic morphologies (i.e.,
sheets, tubes, cones, and icosahedra) and strategies to control their final assembled states. Finally, I will show new machine-learning-enabled coarse-grained models that we have developed to accelerate identification of thermodynamically driven versus kinetically driven morphological transitions during protein self-organization. The insights from these studies reveal the importance of a dynamical perspective on structure-function relationships and highlight the utility of multiscale simulations and statistical inference methods.

Bio: 

Alex is currently an Assistant Professor of Chemical and Biological Engineering at Colorado School of Mines, where he has been since January of 2021. He received his B.S. in Chemical Engineering from M.I.T and his Ph.D. in Chemical Engineering from UT Austin. His graduate research focused on fundamental charge storage mechanisms using carbon-based nanomaterials for supercapacitor applications. As a postdoc, Alex received the F32 NIH Postdoctoral National Research Service Award, which supported his transition into computational biophysics as part of the Chemistry Department at the University of Chicago. Alex is the recipient of the R35 MIRA award from the NIH. His group focuses on the development of multiscale simulation techniques and their application toward both fundamental understanding and engineered control of self-assembled complexes, including for macromolecules, polymers, and porous crystals.