I used to remove all water crystals from pdb file before starting simulation same az what I learned from "lysozyme in water" justine tutorial, but I met some one in a seminar, who told me we must keep water molecules because they have important intractions with special sites in protein and if we remove them, it is like that we remove some parts of the our protein!!!!!!
now I will be grateful to know your ideas
Thank you
In fairness, my tutorial does explicitly say:
"Note that such a procedure is not universally appropriate (i.e., the case of a bound active site water molecule)."
Removing them in the tutorial de-complicates the topology so that there aren't issues later. It's a teaching strategy, and the caveat is stated clearly.
The decision on what to do depends on the biology of the system at hand. Do the waters serve a structural or catalytic role? What were the crystallization conditions? What are the B-factors of those water O atoms (how reliable the positions are)?
In fairness, my tutorial does explicitly say:
"Note that such a procedure is not universally appropriate (i.e., the case of a bound active site water molecule)."
Removing them in the tutorial de-complicates the topology so that there aren't issues later. It's a teaching strategy, and the caveat is stated clearly.
The decision on what to do depends on the biology of the system at hand. Do the waters serve a structural or catalytic role? What were the crystallization conditions? What are the B-factors of those water O atoms (how reliable the positions are)?
I would simulate both systems (with crystalline waters or not), at least for a hundred nanoseconds, and see if there's a difference (If and Only If You have time and computing power)
And probably simulate the system with implicit solvent :)
(Thinking about it, I guess there should be a paper on this...)
And then compare them..something is different?
If you do not have the time and/or the computing power, you could leave the crystalline waters (those with a nice electron density) and add more waters (gromacs) and just minimize the whole system for a longer time...
Hi, I want to ask something quite similar. I'm very new to gromacs and starting my own simulation really confuse me.
In my protein, the water serve a catalytic role with the ligand. That being said, does it means that I shouldn't remove the water molecules at all? And if I don't remove any water molecule, is it still necessary to use any water model?
Thank you
The safest bet is to retain crystal waters. If they're strongly bound or important, they'll stay there. If not, they'll just diffuse like anything else. See the suggestions I made above.
The water model is not really a free choice. Biomolecular force fields are balanced against specific water models and as such should be viewed as a part of the force field.
Not all water, please header of pdb file, you need fix some crystal waters.
- Badges
- Science topic