[gmx-users] Doubt regarding extracting energy component while pull code is applied.

Tushar Ranjan Moharana tusharranjanmoharana at gmail.com
Mon May 23 16:09:25 CEST 2016


Hi Everyone,
I want to model the substrate specificity of an enzyme (lipase). I have
created a mutant which has specificity toward some fatty acid (long chain
unsaturated). I want to analyze the reason behind it. I have done the
following things please let me know if I am going wrong anywhere. Please
feel free to comment.

I created micelles by adding 100 triglyceride molecules and simulating them
for 10 ns as per the literature. It took spherical shape before 5 ns of
simulation. I placed lid open (active conformation ) of lipase close to the
micelles and apply a pull code (mentioned below) to move one of the
carboxyl carbon of one triglyceride molecule near active serine residue.
(Since the active serine is buried in the active cavity non of the
triglyceride move close to it without the pull code. From the theoretical
turnover calculation it looks like it takes milliseconds) Above thing was
done for both mutant and wild type lipase in exactly the same manner.

I want to monitor the interaction of each active site amino acid with the
triglyceride on which the pull code was applied. Is it fine to extract the
energy component while pull code is applied? Or shall I just exclude the
atoms on which the pull code is applied and find interaction of rest of the
atom with active site amino acids? I am using very slow pull rate (1 nm/ns)
and hoping that the system will be in equilibrium. If that isn't the case
then what will be the best thing to do? I am not interested in binding free
energy calculation instead I want to calculate interactions (vs distance of
substrate and active serine) and if possible the energy barrier that the
substrate has to overcome to reach near active serine.

Pull code:
pull            = umbrella
pull-geometry   = distance  ; simple distance increase
pull-dim        = Y Y Y
pull-start      = yes       ; define initial COM distance > 0
pull-ngroups    = 2
pull-ncoords     =1
pull-coord1-groups = 1 2
pull-group1-name = SER_h
pull-group2-name = LIG_hyd
pull-coord1_rate      = -0.001      ; 0.001 nm per ps = 1 nm per ns
pull-coord1_k         = 1500      ; kJ mol^-1 nm^-2


I am using "charmm36-jun2015" forcefield (as there is polyunsaturated fatty
acid in my triglyceride), tip4p water model other parameters are as
recommended in the concerned paper. topology of triglycerides were
generated by http://www.swissparam.ch/. Ions are added to mimic the
reaction condition.

Thanks a lot for your advice.

"A society with free knowledge is better than a society with free food"

Tushar Ranjan Moharana
B. Tech, NIT Warangal
Ph D Student, CCMB


More information about the gromacs.org_gmx-users mailing list