[phenixbb] refinement problem

Sneha Rangarajan rsneha at umd.edu
Tue Sep 30 08:21:01 PDT 2014


I have a 1.9A dataset for a single chain Fv-peptide complex that shows psedomerohedral twin operators in Xtriage (1 in p212121, 3 in p21 and 7 in p1).

Although most programs (in CCP4) identify the best SG to be p212121, MR cannot find a solution with p212121 due to packing clashes.

I tried reprocessing it in p21 and MR can find only one molecule in the asu. The map looks quite good with good density for the sidechains. However the Rfactors wont come down.
After a few rounds of refinement (rigidbody, Xray, TLS and Ind B fac) it is still 38/48.

I tried processing it in p1 and have managed to find 3 molecules in the asu (missing one molecule). The map looks good. While the Rfactors are lower, the gap between Rwork and Rfree is still high (27/40).

I don't know what space group I should choose to ultimately refine it and get the Rfactors low.

If you need any more information, I would be happy to provide it. Does anyone have any suggestions?

Thanks a lot,

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20140930/9565c0ba/attachment.htm>

More information about the phenixbb mailing list