Take the 2-minute tour ×
Biology Stack Exchange is a question and answer site for biology researchers, academics, and students. It's 100% free, no registration required.

I am attempting to find potential Hydrogen bonds between Hydrogen donors and aromatic ring acceptors. I do this by predicting the location of Hydrogens on residues and then calculating how far these Hydrogens are from aromatic rings. If a certain Hydrogen is <7.0 Angstroms from a certain aromatic ring, then I take it under consideration: I form the N-H vector, which is the vector created by the Hydrogen under question and the Nitrogen in the backbone of the residue that the Hydrogen belongs to. I test that this N-H vector is pointing toward the plane of the aromatic ring, and I also test that the point of intersection between the plane of the aromatic and the N-H vector is within 6 Angstroms of the center of the aromatic ring.

If all of these conditions are met, then I consider it a Hydrogen bond between the Hydrogen and the aromatic ring. However, my data must be incorrect, because I am seeing situations where a Hydrogen is < 1.0 Angstrom from the plane of the aromatic. Atoms should not be getting that close to each other.

I thoroughly tested my method by hand using an example situation where my code identified one of the sidechain Hydrogens on an ASN is 0.3 Angstroms from the plane of the aromatic of a TRP. Unfortunately, I could not find any bugs. You can find a PDF of this verification here.

Any suggestions on how my method might be flawed would be greatly appreciated.

share|improve this question
2  
Constructive criticism is far more useful than a down vote. –  thoughtadvances Feb 24 '12 at 22:49
    
Try looking at the structure in question using a program like Chimera or PyMol, so that you can see how these programs interpret the direction in which the hydrogen should be pointing. –  Thomas Ingalls Feb 24 '12 at 23:58
add comment

1 Answer 1

up vote 6 down vote accepted

if you are using NMR structures you might be making the mistake of using several superimposed structures - it would be nice to develop with x-ray structures at less than 2.0 A resolution for starters.

Some of the models at low resolution can be sloppy, but submitted after 1994 or so will not have any center to center distances as that's when the x-ray structures used molecular models rather than just electron density rigorously and wierd steric violations in the structure should be rare.

Still, this might not be wrong - the hydrogen might be pointing right into the aromatic ring's center. This has been observed quite a bit. In such a case, the atom to ring plane distance might be very small.

You only need to worry if the center to center distance between atoms is less than the vanderWaals radius. I would filter for atom center to center distance and see if you see violations.

share|improve this answer
    
Excellent idea. Indeed, when I test for the distance from the Hydrogen to the center of the aromatic, I do not find any situations where the Hydrogen is within 1.0 Angstrom. Indeed, situations where the Hydrogen is within 2.0 Angstroms is rare. As for the NMR consideration, I did take care of that. My parser successfully parses all of the NMR's MODEL's, and then it throws out all but the first one for the sake of this project –  thoughtadvances Feb 25 '12 at 6:19
    
Now, I am left with only one problem: I expected to see most Hydrogen bonding situations between aromatic rings and sidechain Hydrogen donors. To the contrary, I am seeing far more Hydrogen bonding situations between mainchain Hydrogens and the aromatic ring. Can you think of any reason why? I would expect the large backbone structure to be unable to get near to the small aromatic rings. –  thoughtadvances Feb 25 '12 at 6:24
1  
Glad you had already worked through a lot of this. I would tend to think that main chain h-bonds will be selected for more strongly as the sidechains tend to change very easily. Sidechains also tend to form stronger (i think stronger) aro-aro or pi-aro interactions and so there may be a tendency for important h-bonds to not from as readily. Selection arguments are hand wavy, but maybe you can think of some clever stat test for this. –  shigeta Feb 26 '12 at 6:20
    
Interesting ideas. Thank you. –  thoughtadvances Feb 26 '12 at 19:33
add comment

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.