Yes, you are right. It only reported positive value. Now I see why HBDA is used in refinement stage, because the hydrogen bond pair is already supposed to be close. HBDA is mainly used to control angle, not distance. That's impressive. Thanks a lot!
On Fri, Nov 4, 2016 at 8:59 AM, Charles Schwieters <[email protected]> wrote: > > Hello Qinhong-- > > > > > > > > > Does HBDA only work in a limited distance? I have this question > because a > > > > hydrogen bond pair defined in my HBDA table is always far away (>6A) > from > > > each > > > > other, while it is not reported in violations. Other hydrogen bond > pairs > > > close > > > > from each other are normally reported if violated. So I think maybe > the > > > software > > > > thinks there's no hydrogen bond at all when exceeding certain > distance, and > > > thus > > > > not using that HBDA term? > > > > > > > > > > The HBDA term is not a distance restraint. Rather, it enforces a > geometrical > > > relationship between the HN-O distance and the O-HN-N angle. A > > > distance restraint is also required to keep the atoms close. Please > > > see J. Am. Chem. Soc. 124, 10621-10626 (2002). > > > > Yes, I know HBDA is a distance-angle relation restraint rather than > > distance restraint. The problem is there's a pair with N-H...O angle > > 107 degree while H...O distance is 7.6A, which is obviously violated > > HBDA restraint, but not reported in violations. I think that term > > may be not used at all because they are too far away. > > There is no distance cutoff in the HBDA term. This is the expression > in the hbda code: > > (1./(RHO**3))-0.019-(0.21/(2.07+ COS(THETA))**3) > > where RHO is the distance (in angstroms) and THETA the angle in > radians. I believe the result is negative, and thus will not be > printed in the violation analysis. > > > By the way, > > that hydrogen bond is between protein and ligand, does it matter? > > If it's in the HDBA restraint table it shouldn't matter. > > best regards-- > Charles > -- Qinhong Yu Ames' Lab Department of Chemistry UC Davis
_______________________________________________ Xplor-nih mailing list [email protected] https://dcb.cit.nih.gov/mailman/listinfo/xplor-nih
