Deep learning

I'm using tools and algorithms borrowed from recent work in deep learning to build better models for protein dynamics.

A side effect of the recent boom in deep learning has been the proliferation of well-engineered libraries for performing arbitrary symbolic differentiation. I’m using these tools to enrich our current dynamical modeling techniques.

In addition, I’m investigating the use of deep networks to find better feature representations of biomolecules.

Ion channels

Sodium channels initiate signaling in neurons and other cells which go on to become sensations of pleasure and pain, as well as thoughts and feelings. Potassium channels return excitable cells to normalcy.

In collaboration with the duBois lab and with the help of Folding@Home, we are conducting large scale molecular dynamics study of a voltage gated sodium channel to compliment experimental work to probe sodium channel function through the use of natural toxins secreted by frogs in the Amazon rain forest.

In collaboration with Pfizer and Folding@Home, we studied a mechanosensative potassium channel transitioning between stretched and compressed conformations.

Markov Modeling

Our lab develops and applies novel statistical techniques to understand and interpret the huge volume of data returned from a molecular dynamics study.

Particularly, we use Markov state models (MSMs) and time-structure-based independent component analysis (tICA) to draw interpretable conclusions from large timeseries data sets.

I’ve introduced a new method for including solvent degrees of freedom in MSM analysis described in this paper.

I’ve introduced a computationally tractable form of non-linear tICA that is analogous to an MSM with soft states here.


  1. Conserve Water: A Method for the Analysis of Solvent in Molecular Dynamics
    Matthew P. Harrigan; Diwakar Shukla; Vijay S. Pande
    February 12, 2015
    J. Chem. Theory Comput., 2015, 11 (3), pp 1094--1101. doi:10.1021/ct5010017 [pdf]
  2. MDTraj: A Modern Open Library for the Analysis of Molecular Dynamics Trajectories
    Robert T. McGibbon; Kyle A. Beauchamp; Matthew P. Harrigan; Christoph Klein; Jason M. Swails; Carlos X. Hernández; Christian R. Schwantes; Lee-Ping Wang; Thomas J. Lane; Vijay S. Pande
    October 20, 2015
    Biophys. J., 109, 8, pp 1528--1532. doi:10.1016/j.bpj.2015.08.015 [bioRxiv] [pdf]
  3. User-friendly software for analyzing MD simulations
    December 01, 2015
    Nature Methods 12, 1116. doi:10.1038/nmeth.3669
  4. Osprey: Hyperparameter Optimization for Machine Learning
    Robert T. McGibbon; Carlos X. Hernández; Matthew P. Harrigan; Steven Kearnes; Mohammad M. Sultan; Stanislaw Jastrzebski; Brooke E. Husic; Vijay S. Pande
    September 07, 2016
    JOSS, 1, 5. doi:10.21105/joss.00034
  5. MSMBuilder: Statistical Models for Biomolecular Dynamics
    Matthew P. Harrigan; Mohammad M. Sultan; Carlos Xavier Hernandez; Brooke E. Husic; Peter Eastman; Christian R. Schwantes; Kyle A. Beauchamp; Robert T. McGibbon; Vijay S. Pande
    January 10, 2017
    Biophys. J., 112, 10--15. doi:10.1016/j.bpj.2016.10.042 [bioRxiv]
  6. OpenMM 7: Rapid Development of High Performance Algorithms for Molecular Dynamics
    Peter Eastman; Jason Swails; John D. Chodera; Robert T. McGibbon; Yutong Zhao; Kyle A. Beauchamp; Lee-Ping Wang; Andrew C. Simmonett; Matthew P. Harrigan; Bernard R. Brooks; Vijay S. Pande
    December 06, 2016
    bioRxiv. doi:10.1101/091801
  7. Markov modeling reveals novel intracellular modulation of the human TREK-2 selectivity filter
    Matthew P. Harrigan; Keri A. McKiernan; Veerabahu Shanmugasundaram; Rajiah Aldrin Denny; Vijay S. Pande
    April 04, 2017
    Scientific Reports, 7, 632. doi:10.1038/s41598-017-00256-y [bioRxiv] [medium]
  8. Landmark Kernel tICA For Conformational Dynamics
    Matthew P. Harrigan; Vijay S. Pande
    April 04, 2017
    bioRxiv. doi:10.1101/123752