Hi,
I would like to study how a large protein complex interracts with a ligand. Due to the sheer size im afraid im limited to Coarse Grained.
Im guite confident about the protein, however the ligand scare me a bit. I found this paper Article Automated Parametrization of the Coarse-Grained Martini Forc...
however to low number of citations does not convince me.Could anybody comment a bit on my methodology. Does it make any sense to try to parametrize the ligand in CG?
best
T
Is it important for you to include the entire protein complex in the simulation? If not, run atomistic MD simulations: define the simulated system by the position and size of the non-periodic solvent sphere.
Hi Tomek Stępniewski
If you have to cope with a limited computational efficiency, the coarse-grained force-fields could solve your problem.
If you want to work with MARTINI force field, you should do the following steps:
- Ligand parametrization. This procedure will requires several steps and experimental validation through partition coefficient calculation.
The partition coefficient of your ligand could be easily calculated through CG-MD simulations thermodynamic integration approach.
Tutorial:
http://cgmartini.nl/index.php/tutorials-general-introduction/parametrzining-new-molecule
-Protein parametrization. You have to assign secondary structure parameters.
Pay attention to the fact that the secondary structures in the binding site can not change.
This procedure can be easily conducted as reported in the following tutorial:
http://cgmartini.nl/index.php/tutorials-general-introduction/proteins.
An interesting alternative consists in the use of hybrid force-field (MARTIN/ All-atoms). The use of hybrid force-field will allow to monitor the hydrogen-bond dynamics and will allow to investigate the secondary structure transitions occuring in the binding site. This approach overcomes the main limitation of MARTINI force-field, i.e. the use of fixed secondary structure parameters.
In addition, this approach allows to reduce the computational costs. Indeed, the solvent can be modeled according to the CG-description, while the ligand and the binding site according to the UA-description.
I hope this answer can clarify some doubts.
- Badges
- Science topic