Dear Rohit, You can follow the way Tim suggested. At low resolution and for disordered regions, autobuster works well for refinement.
Atul On Tue, Jan 13, 2015 at 7:47 AM, Tim Gruene <[email protected]> wrote: > Dear Rohit Kumar Singh, > > you can remove a couple of residues before and after the outlier and > rebuild it. You can also switch on Ramachandran restraints in Coot and > run real-space refinement. If this does not help, I would delete a > single residue within the region in question, run real-space refinement > for both segments and reconnect. > > Regards, > Tim > > On 01/13/2015 08:35 AM, rohit kumar wrote: > > Hello all, > > > > if outlier is in between 5 to 6 %, how someone can fix it. > > As the resolution is low (3.0-3.5 A). > > > > On Mon, Jan 12, 2015 at 7:53 PM, Robbie P. Joosten <[email protected]> > wrote: > > > >> Hi Dialing, > >> > >> 86% favoured can actually be quite okay for an initial model. As long as > >> it doesn't have 14% in the disallowed region, the vast majority of > residues > >> will be correct, or close enough to the correct answer that they can be > >> fixed easily. Of course, such a model will need tweaking, but this is > more > >> or less the point of calling it an 'initial' model. > >> > >> Cheers, > >> Robbie > >> > >> > >> On 01/12/2015 01:18 PM, Dialing Pretty wrote: > >> > >>> Dear All， > >>> > >>> If an initial PDB has only 86% residues in the Ramachandran favored > >>> region, it > >>> would mean there is a significant error (for example significant > length of > >>> protein fragment in the total protein assigned to the wrong electron > >>> density map > >>> position) , right? > >>> > >>> Dialing > >>> > >>> > > > > > > -- > Dr Tim Gruene > Institut fuer anorganische Chemie > Tammannstr. 4 > D-37077 Goettingen > > GPG Key ID = A46BEE1A > >
