I notice that during manual rebuilding cis peptides are frequently put in by mistake. It is considerable manual work to make them trans and would be nice to follow that with torsion dynamics with say a 3.1 A structure we are working on. Coot can find a list of cis peptides and it is often a surprise to postdocs to see it. I will ask mehmet to send you structure and list of the non-pro ones. Sent from my iPhone On Jan 30, 2011, at 12:52 PM, Nathaniel Echols wrote:

On Sun, Jan 30, 2011 at 8:26 AM, Timothy Springer wrote:

...

Is there an easy way to specify unwanted cis-peptides and remove them in refinement or with torsion dynamics or SA?

Aside from Ralf's suggestions, there isn't currently a way to do this explicitly. My understanding is that SA (torsion or Cartesian) usually isn't sufficient to rotate a peptide 180 degrees, although I guess it's worth a try. However, I'm working on code to automatically correct misfit peptides, and it wouldn't be difficult to specifically handle cis-peptides, and to accept an explicit atom selection instead of looping over all residues. (It's only semi-functional right now - it runs, but doesn't always improve the model; fixing it is on my agenda for the coming week.) If you're willing to share the model and data that you're concerned about, it might speed up my testing.

thanks, Nat _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb