I've seen a similar behavior when, for the Phenix.refine refinement
strategy, I accidentally selected both XYZ (reciprocal-space) and XYZ
(real-space) at the same time. As well as a Phe jumping out of density
in one NCS copy, planarity was very poor for several sidechains (but
no obvious pattern as to which ended up with poor geometry).
Hope this helps,
Andy Purkiss.
Quoting "Philip D. Jeffrey" <[email protected]>:
As Paul observes, there's a bona fide email list for phenix, but in
fact comparing REFMAC to phenix.refine might be useful to see if
this really is a program bug or if it's a bug in the model.
Sometimes one is better than the other.
For phenix.refine:
Put the Tyr in the "right" place and scour the .geo file created at
the start of refinement for bad angles or bonds or bumps etc. This
will tell you what the offending energy term is, if there is one.
If not, try turning off the rotamer fixing aspect of phenix.refine.
I've had aspects of the algorithm "anti-refine" a structure
(geometry, Rwork, Rfree all got worse at once) so I tend to
streamline the process and run it as:
strategy=individual_sites+individual_adp+occupancies (+tls if
you're using it)
from the command line (since I'm a Luddite and mostly eschew the
GUI). For example if it's enforcing rotamers during NCS refinement
it could pull your s/chain out of density.
But also try REFMAC if the problem persists.
Cheers
Phil Jeffrey
Princeton
________________________________
From: CCP4 bulletin board <[email protected]> on behalf of Paul
Emsley <[email protected]>
Sent: Monday, October 19, 2020 2:58 PM
To: [email protected] <[email protected]>
Subject: Re: [ccp4bb] Tyr pushed out during refinement
On 19/10/2020 19:52, Boniecki, Michal wrote:
I have a problem during refining with one of the Tyr residues. It is
constantly pushed out of the position during refinement in all 4
chains in ASU.
I have tried to exclude it from refinement in phenix but it is
refined anyway out of the position with wrong geometry. Is there a
possibility to fix it during refinement?
If this is a question about refinement in phenix, then this may not
be the most useful mailing list. If this is a question about how to
pin down atoms and tweak bonds and contacts in Coot or Refmac, then
we can proceed.
Regards,
Paul.
________________________________
To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
########################################################################
To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a
mailing list hosted by www.jiscmail.ac.uk, terms & conditions are
available at https://www.jiscmail.ac.uk/policyandsecurity/
########################################################################
To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list
hosted by www.jiscmail.ac.uk, terms & conditions are available at
https://www.jiscmail.ac.uk/policyandsecurity/
