Python tools for NAMD
PyNAMD is a package meant to provide a flexible, lightweight interface to NAnoscale Molecular Dynamics (NAMD) input and output. Since many powerful and mature packages exist for trajectory analysis, the focus is almost exclusively on energy based output.
The PyNAMD library is also accompanied by several scripts for common tasks useful in molecular dynamics (MD) simulations, such as rapidly computing averages and fluctuations - all directly from NAMD output. Developments are ongoing to provide considerably more complicated analysis tools such as multistate reweighting methods (e.g., WHAM) for generalized ensemble, replica exchange, and stratified umbrella sampling simulations.
To install PyNAMD, clone this git repository and change into the new "pynamd" directory. PyNAMD can then be built with the command
python setup.py install
If you are using the Python distribution that comes with your operating system, you may need to run the above command with administrative privileges or else add the --user flag -- the latter is recommended. It is expected that this project will utilize a considerable amount of "bleeding edge" numerical tools, so it would also be a good idea to use a customizable Python environment that has the latest numpy and scipy, such as that provided by Anaconda.
import pynamd
log = pynamd.NamdLog("00001.log", "00002.log")
#info contains information about time-steps, temperatures etc...
log.info
#energy contains the time-series data in a dictionary
log.energy
#log.energy can easily be converted into a pandas dataframe
import pandas as pd
df = pd.DataFrame(log.energy)Further examples are in the examples subfolder.
Tests can be run using the pytest framework. Install pytest using conda conda install pytest or pip pip install pytest.
Run with pytest in the root of the repository.
- Brian Radak | [email protected] | [email protected]