Posts

Showing posts from December, 2015

Native contact (Q) biasing in GROMACS and plumed2

Fraction of native contacts is a pretty darn good coordinate for looking at protein folding, definitely when it comes to small proteins. Just have a look:

http://www.pnas.org/content/110/44/17874.abstract

Fraction of native contacts can be used to analyze long, unbiased MD runs – like those from DESRES – or in biasing simulations. But how does it fare when used in biasing? Well I don't know – but you can now find out by yourself! Either in umbrella sampling or metadynamics – or a method of your choice that doesn't use energy as your collective variable.

Here, I'm happy to report that I incorporated an implementation of native contact CV into plumed2 – an enhanced sampling package that can be used with a number of popular MD simulation engines. The pull request is now closed https://github.com/plumed/plumed2/pull/177 and it's likely that the next release of plumed2 will include this.

"Tutorial" is a regression test in plumed2 sources  regtest/basic/rt29/ which …

Scripting paramchem.org access

Below is a simple python script that allows one to programatically access paramchem.org, a small-molecule parametrization service. Autocorrect said "problematic" instead of "programatic", maybe I'm noting getting the hint...

The script has minimal dependencies and is very "light-weight", meaning no error handling or error messages. A full working example for a benzene (duh!) is attached.

Long time ago, I had to implement something similar to this script, so this was a fairly easy job. It relies on mechanize and beautiful soup to do it's thing.

python paramchem.py -u "YOUR USERNAME" -p "YOUR PASSWORD" -c benzene.pdb

# manually correct the "RESI" line in benzene.str to be 
# RESI BNZ 0.000 ! param...

python cgenff_charmm2gmx.py BNZ benzene.mol2 benzene.str charmm36-jun2015.ff/

Source code and examples
git clone https://github.com/jandom/paramchem
Example on google drive

Edited: note on errors, link to github repo