[phenixbb] molprobity vs validation_cryoem

[Oleg Sobolev]

Hi Roberto,

The difference occurs because of using/not using CDL
(conformation-dependent library) modifications. In case of molprobity, CDL
is not used since it is not mentioned the pdb file header which lead us to
assume that it was refined without CDL. In case of cryoem_validation it is
used. To turn it off use "cdl=False" command-line parameter or locate this
flag in GUI. To turn it on for molprobity, add line
REMARK   3 CDL
to your pdb file.

Best regards,
Oleg Sobolev.

On Fri, Apr 26, 2019 at 6:14 AM Roberto Marabini <roberto at cnb.csic.es>
wrote:

> Sorry, In the previous email I sent a wrong link for the pdb file. The
> rigth one is:
>
>
> https://drive.google.com/open?id=0B_i_LJfUN1z7dEF6MHhzNEJvUEo2eUJCZmNRYUpfVFFLZy1j
>
> On Fri, Apr 26, 2019 at 10:51 AM Roberto Marabini <roberto at cnb.csic.es>
> wrote:
>
>> Hi guys,
>>
>>    We are using phenix version 1.15.2-3472. We have executed molprobity
>> using the command line:
>>
>> phenix.python_1.15.2-3472
>> /usr/local/phenix-1.15.2-3472/modules/cctbx_project/mmtbx/command_line/molprobity.py
>> "path_to_pdb"/refmac-refined.pdb pickle=True
>> nproc=4
>>
>> We have also executed validation_cryoem (here we have used both the GUI
>> and the command line and both interfaces return the same result)
>>
>> phenix.python_1.15.2-3472
>> ./phenix/phenix/command_line/validation_cryoem.py
>> "path_to_pdb"/refmac-refined.pdb "path_to_map"/molprobity.mrc
>> resolution=3.2 pickle=True
>>
>> What we do not fully understand is why the two commands return different
>> results for molprobity. Are the default parameters different?
>>
>> For example, validation_cryoem reports as number of outliers "34" in the
>> molprobity tab (model.geometry.angle.outliers) while molprobity reports 21
>>
>> The test files are available at:
>>
>> 3Dmap:
>> https://drive.google.com/open?id=0B_i_LJfUN1z7amoyM1B2N1U0Y2VYODB6LWdLY1pwai04RWxz
>> pdb file: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5093775/
>>
>>     Thanks for the help
>>
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb
> Unsubscribe: phenixbb-leave at phenix-online.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20190426/c35cb5e9/attachment.htm>