[gmx-users] A question about Protein-Protein Docking with Gromacs

[milad bagheri]

In many cellular signaling pathways in cancer cells, the expression of many
genes rises. One result of this excessive expression is cellular traffic
and unwanted interactions between proteins. Therefore, it is important to
examine the interactions and identify the amino acids involved in the
interactions.
There are several ways to examine the interactions such as string and using
molecular docking.
Today, many Docking softwares are active on web servers like Haddock,
cluspro, zdock, etc; but, to my mind, examining the interaction between
proteins using docking, based on several different reasons, has low
credibility that makes the results of these softwares unreliable.
That's why I decided to examine it with a better method. For this purpose,
I used the molecular dynamics simulation method to locate the molecules
under dynamic conditions to make it easier to study the interactions
between amino acids (fluctuations, pi pi, hydrogen, and electrostatic
bonds).
The procedure is as follows:
1- Two proteins were received from the RCSB site.
2- Two proteins were placed at 6-angstrom distance using the Discover
Studio software. First question is: Please suggest another way to do this
procedure via software (considering the desired position) (according to the
available data in the literature) (i.e. the hot spot amino acids) (involved
in the interaction) to place two proteins of appropriate position in front
of each other? And the next question is whether the 6-angstrom gap between
two proteins is sufficient?
3- From the topology building process with pdb2gmx to indexing like the
rest of the Justin Lamquell site tutorials were done (Multiple chain
tutorial). But the problem is choosing the size of the box to reduce the
simulation time. Please suggest a way to find the appropriate box size for
two proteins in a box.
4- In order to force the interaction of two proteins, the Pull code
tutorial was used.
It should be noted that a protein was used to capture RMSD; i.e. the
protein stabilization was screened.
Please suggest a more precise way to analyze, develop, and validate this
method.