Number of contacts between ligand and enzyme - hydrogens yes or no?

User discussions
1

GROMACS version: 2020
GROMACS modification: Yes

Hello,

I’ve seen examples where the number of contacts between a ligand and enzyme is determined. Is this calculated with or without hydrogens typically? I’ve seen examples where the average # of contacts per ligand heavy atom is quoted, which implies without. I was curious if anyone here knows which is more common (if one is)

Thanks in advance
Sam

2

I would never use hydrogens for this kind of calculation. Experimentally difficult to determine and it basically just inflates the number of contacts (is a CH3 really going to contribute 4 contacts to the count when it would suffice to use the C atom only?).

3

I agree. Do you use the -group option?

I have a pair of drugs that are identical other than one has a benzene ring and one is anthracene, both binding to the same residue (arginine). My understanding is that -group would count interactions between said arginine and benzene/arginine and anthracene as 1, whereas without -group, anthracene has more heavy atoms for potential contacts etc.

Again, it could just superfluously inflate the number of contacts. I suppose you could argue that not using -group would demonstrate the difference between the two moieties better than -group would.

Thank you for your advice as always
Sam

4

I don’t see why one would want to in most cases. If you want the number of contacts between two groups, simply constructing appropriate index groups (without H atoms) and choosing them both when prompted from gmx mindist -on should do the job.

5

Hello,
I am new in your community and not sure how to connect with the appropriate group!

Hence, here is my question:
Which command would calculate the number of contacts between two identical proteins (homodimer), more specifically “per residue”?

Thanks
Roger

6

You need to set up appropriate index groups to do that, in concert with gmx mindist -on. No command decomposes this quantity per residue, you’d have to make individual index groups, one for each residue of interest, and perform the analysis.

7

I apologize for the late response on this topic. What type of contacts would be calculated in such case? If we exclude H-atoms, does it mean that gmx mindist will exclude all possible H-bonds from the calculation? If yes, does it also mean that we’ll be able to conclude about non-polar contacts after such calculation?