scipy2015-mdsynthesis: MDSynthesis: a Python package enabling data-driven molecular dynamics research

Proteins are the molecular machines of life, and molecular dynamics (MD) simulation permits atomically-detailed access to the interactions that define their behavior. MD trajectories give a sequence of structural snapshots--a timeseries of thousands to many millions of atom positions. With ever-faster computational resources we can routinely generate many terabytes of data, perhaps spread over hundreds of individual simulation trajectories. Often these trajectories are not just repeats of the same simulation system, but instead sample a wide range of different starting configurations, forcefield parameters, protein conformations, mutations, protonation states, etc., which can make data management difficult. Furthermore, because of their cost to calculate, it is necessary to store intermediate data--often timeseries for specific structural or thermodynamic quantities of interest. This adds to the complexity of managing data, and serves as a barrier to answering scientific questions. MDSynthesis, a Python package that handles the tedious and time-consuming logistics of intermediate data storage and retrieval, is in active development to address this problem. MDSynthesis features container objects that use the robust MDAnalysis library for dissecting the details of individual MD trajectories, and they store their states to disk on-the-fly using the PyTables HDF5 interface. These containers are memory efficient and built for aggregation, including convenience methods for quickly combining and comparing datasets (pandas, numpy, or other pure python structures) across hundreds of simulations in arbitrary ways. This makes data exploration feasible with MD data, and the abstraction the containers provide make it easier to write analysis code that works across many variants of a simulation system. The package is actively developed and freely available under the GPLv2 from https://github.com/Becksteinlab/MDSynthesis.

This poster was presented on 2015.07.08 at SciPy 2015 in Austin, TX.