Hi, You're right that the left-handed helix is a small minimum in the Ramachandran plot.
In calculating structures with limited experimental data + repulsive vdWs, you do see an unusually large number of residues in the left-handed helix. That's largely caused by the relative inaccuracy of the repulsive vdW potential. The best way to correct this is to use xplor-nih's RAMA potential in addition to the experimental data & repulsive vdWs. The RAMA potential pushes rotatable bonds toward physically reasonable combinations, with lower energies for more-commonly seen combinations of torsion angles (i.e., the right-handed helix has a much lower RAMA energy than the left-handed helix). Examples of its use are in xplor/eginput/gb1_rdc/anneal.py (for Python) xplor/eginput/marvin/cvn/sa_pass2.tcl (for TCL) xplor/eginput/fbp/domain2_tor_final.inp (for classic xplor) If you want background information, I had a couple of papers describing the technique: Prot Sci 5: 1067 (1996) J Magn Resn 125: 171 (1997) J Magn Resn 146: 249 (2000) J Am Chem Soc 124: 2866 (2002) Hope this helps. --JK On Apr 18, 2007, at 10:31 AM, Shi,Lei wrote: > Hi groups, > I am using Xplor-nih to refine a small protein with two helical > domains > and a loop connecting them. I am using NOE restaints and torsion angle > data. > When I analyze the final structure, I found that there are 1-2 > residues > in the left hand helix region in the Ramachandran plot for a > overall of > around ~50 structure. I use torsion angle dynamics and only use the > replusive vdw potential. > I do not have a lot of restraints in the loop region. I do not know > whether it is the reason I have those left handed helix. It is rare > to see > the left handed though. > I do not know what I might have done wrong. Any help is greatly > appreciated. > Thanks, > Lei > > _______________________________________________ > Xplor-nih mailing list > Xplor-nih at nmr.cit.nih.gov > http://dcb.cit.nih.gov/mailman/listinfo/xplor-nih
