[Chimera-users] H Bonds in MD simulation

Elaine Meng meng at cgl.ucsf.edu
Thu Jan 28 14:53:14 PST 2010

Different programs use different methods and cutoffs.  You could try  
setting the tolerances to 0.0 if you feel the default Chimera method  
is too permissive, but there is no reason to expect exactly the same  
number of H-bonds from two different programs.  You would need to look  
at VMD documentation to see how theirs are calculated, and at  
Chimera's (link already sent in previous message) to know how they are  
calculated, then decide which one you want.

In Chimera, the angle and distance cutoffs depend on the atom types,  
and the criteria depend on published data (the small molecule  
structures).  In many other programs, the same cutoffs are used no  
matter which atom types are involved.  I don't know how it is done in  
VMD, however.

I know of one paper where the authors tried both MolProbity (a fancier  
method) and Chimera and said they usually gave the identical H-bonds:
Differences in the risk of celiac disease associated with HLA-DQ2.5 or  
HLA-DQ2.2 are related to sustained gluten antigen presentation.
Fallang LE, Bergseng E, Hotta K, Berg-Larsen A, Kim CY, Sollid LM.
Nat Immunol. 2009 Oct;10(10):1096-101


On Jan 28, 2010, at 2:36 PM, snoze pa wrote:

> Hi Elaine, Thanks for your help. But somehow I am getting different
> number of bonds in vmd and chimera. A snap shot of the  figure is
> shown below.
> http://img191.imageshack.us/img191/9084/screenshotva.png
> In VMD the Hbonds are calculated by show Hbonds and protein. Vms shows
> 24 hbonds while chimera shows 38 hbonds. I don't know how to justify
> this. I will highly appreciate your help. A snapshot is shown in the
> link
> http://img191.imageshack.us/img191/9084/screenshotva.png
> Thanks
> s

More information about the Chimera-users mailing list