Hi Kostya, I have seen worse data than that, which are ususally treated quite well within phenix using standard parameters. With a resolution of about 2.7 Ang. I wouldn't expect to see waters and also just well ordered side chains, whereas more flexible ones just give some "bumps" pointing away from the main chain density. Though without a picture or something else it is difficult to say if your maps are really unusually bad. Cheers Christian Am 05.10.2016 um 07:47 schrieb Kogan, Konstantin:

Dear all,

I have highly anisotropic data. I was able to solve the structure by MR, and refine it with phenix.refine to Rwork/Rfree=0.28/0.31 with Resolution cutoff 2.64Å and mean B values 114, which is pretty high for such resolution. Though the overall structure makes sense, the maps are quite featureless, no water molecules, and side-chains refinement is a problem. I have tested how the data is anisotropic with Diffraction Anisotropy Server https://services.mbi.ucla.edu/anisoscale/ (see attached image), which shows clearly, that we don't really have data to 2.64Å in all directions. Though, there are other servers/programs that can deal specifically with anisotropic data e.g. STARANISO anisotropy server http://staraniso.globalphasing.org/cgi-bin/staraniso.cgi and then refine with BUSTER, I would like to know if it's possible still to use Phenix with some more advance parameters to actually address the issue of anisotropy and to get maximum out of the data. The space group is P61 2 2, cell parameters are 73, 73, 453, 90, 90, 120 and I have high multiplicity data, but no NCS.

Thanks in advance for any input,

Kostya -- Konstantin (Kostya) Kogan Postdoctoral researcher Pekka Lappalainen's Lab Institute of Biotechnology University of Helsinki Helsinki, Finland Mobile: +358-(0)45-8994342

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

Hi Kostya, how complete the data set is? Missing reflections in some resolution zones between 2.64A and inf may be sufficient to make the effective resolution lower or much lower than 2.64A which would explain why the map does not look like what you expect. Pavel On 10/4/16 23:47, Kogan, Konstantin wrote:

Dear all,

I have highly anisotropic data. I was able to solve the structure by MR, and refine it with phenix.refine to Rwork/Rfree=0.28/0.31 with Resolution cutoff 2.64Å and mean B values 114, which is pretty high for such resolution. Though the overall structure makes sense, the maps are quite featureless, no water molecules, and side-chains refinement is a problem. I have tested how the data is anisotropic with Diffraction Anisotropy Server https://services.mbi.ucla.edu/anisoscale/ (see attached image), which shows clearly, that we don't really have data to 2.64Å in all directions. Though, there are other servers/programs that can deal specifically with anisotropic data e.g. STARANISO anisotropy server http://staraniso.globalphasing.org/cgi-bin/staraniso.cgi and then refine with BUSTER, I would like to know if it's possible still to use Phenix with some more advance parameters to actually address the issue of anisotropy and to get maximum out of the data. The space group is P61 2 2, cell parameters are 73, 73, 453, 90, 90, 120 and I have high multiplicity data, but no NCS.

Thanks in advance for any input,

Kostya