+1
> -----Original Message----- > From: Mailing list for users of COOT Crystallographic Software > [mailto:[email protected]] On Behalf Of Dirk Kostrewa > Sent: Thursday, April 14, 2011 5:37 AM > To: [email protected] > Subject: Re: [COOT] low res real-space refinement > > Another idea, that I propose from time to time, would be to have a > complete force-field like geometry term, including VDW repulsions and > hydrogen bonds. This would avoid refining atoms into already occupied > places and it would maintain secondary structures even at very low > resolution. > > Best regards, > > Dirk. > > Am 14.04.11 11:32, schrieb Bernhard Lohkamp: > > This is an idea I had for a while but never had the time to implement > > or test (it's still on my list though). For now torsion angle > > refinement and especially back-rub rotamer fitting of side chains > > (default for lower resolution; 2.8A IIRC) should help to prevent side > > chains from moving into the 'main' density. > > > > B > > > >> this one is for Paul, but maybe it's worth discussing it on the > list. > >> If that > >> feature already exists, please let me know. > >> > >> Would it make sense to down-weight the side chain atoms in real > space > >> refinement? At low resolution I often find that the side chain atoms > >> are dragged > >> into the density of where I suspect the main chain to run, and maybe > >> this effect > >> could be avoided by downweighting the X-ray 'force' on the side > chain > >> atoms. > >> This might correspond to having different matrix values for main and > >> side chain > >> atoms. > >> > >> I am curious to hear comments about this idea. > >> > >> Cheers, Tim > >> > > > > > > -- > > ******************************************************* > Dirk Kostrewa > Gene Center Munich, A5.07 > Department of Biochemistry > Ludwig-Maximilians-Universität München > Feodor-Lynen-Str. 25 > D-81377 Munich > Germany > Phone: +49-89-2180-76845 > Fax: +49-89-2180-76999 > E-mail: [email protected] > WWW: www.genzentrum.lmu.de > *******************************************************
