Hi John, I want the molecule to do the rotations, since I oriented the helices in another fashion. Subsequently I introduced an alignment tensor which is oriented in such a way that helix RDCs are optimized.
Then I introduce the other RDCs, and want that part of the molecule to find it's orientation matching the current alignment tensor. For clarity: I am working on a nucleic acid element which consists of two helices connnected by a joint. This joint part poses the problem. Kind regards, Ramon On Mon, 2008-03-03 at 10:37 -0500, John Kuszewski wrote: > Hi Ramon, > > Looks like your contraints fix statements are the problem. By fixing > the entire SANI "molecule," you've forced your structure to do all > the overall rotations necessary for RDC agreement. But by fixing > parts of your structure as well, you've prevented that. > > Change the first constraints fix selection to (segid SANI and resid > 500 and name OO), and get rid of the selections for parts of your > protein, and you should be all right. > > But note that you'd be better off using a more modern script, like > xplor/eginput/gb1_rdc/refine.py to do this. > > Hope this helps. > > --JK > > On Mar 3, 2008, at 5:12 AM, R.M. van der Werf wrote: > > > Hi, > > > > I am trying to do a global RDC refinement, but the molecule appears to > > move very little, or not at all. > > > > My script goes as follows (Temperature = 300K over entire script) > > > > 1. Read structure, parameters, restraints (NOE, RDCs, dihedrals) > > 2. Introduce alignment tensor (pseudoatoms) > > 3. fix helices: > > constraint fix=((segid SANI and resid 500) > > or (segid EDHB and resid 1:5) or (segid EDHB and resid 33:37) or > > (segid EDHB and resid 14:29)) end > > > > My structure contains 37 residues, RDCs need to be refined for the > > residues which are not fixed. > > > > 4. 4000 verlet dynamic steps of 0.0005 ps with the > > flags bonds angle impr vdw noe cdih plan sani > > 5. Loop of verlet dynamics in which in every loop step the NOE and > > cdih > > force is increased > > 6. 20000 verlet dynamic steps of 0.0005 ps, at 'high' force for NOE > > and cdih > > 7. powell minimisation of 10000 steps > > > > Any ideas are welcome. > > Increasing the timesteps moves the final structures to a higher energy > > level (and thus not ending up in an energetic minimum). > > Is another temperature needed ? > > > > Kind regards, > > Ramon > > > > _______________________________________________ > > Xplor-nih mailing list > > Xplor-nih at nmr.cit.nih.gov > > http://dcb.cit.nih.gov/mailman/listinfo/xplor-nih >
