Hi Oli,
ok, great! Please send me an example once you have one and I will investigate.
You can disable it.. The option is local_grid_search. If you use GUI just uncheck that box. Or remove "*" in front of it if you use parameter file. Or don't include it into "run=..." if you use command line.
Note that will also disable using rotamer restraints which may
result in elevated counts of rotamer outliers.
Pavel
Hi Pavel,
I will definitely send an example the next time I come across one. There is no way to turn this off then? In most cases it is very helpful, but towards the end of the model building process I don't necessarily want phenix to alter rotamers that I have manually inspected. Yes, I am using dev-4158, the latest nightly available as far as I can see.
CheersOli
On Jun 11, 2021, at 2:24 AM, Pavel Afonine <[email protected]> wrote:
Hi Oliver,
rotamer fitting is designed to do better or leave things alone untouched. If your experience is different, I'd appreciate seeing an example (so that I can make it better!).
Also, I assume you are using one of latest nightly builds..
Pavel
On 6/10/21 12:41, Oliver Clarke wrote:
From the phenix docs, I can see how to modify this behaviour, but not how to switch it off without just switching to rigid body fitting:
- rotamers
- fit�= all *outliers_or_poormap outliers_and_poormap outliers poormap�Fit side-chains regimes
- tuneup�= outliers outliers_and_poormap�Switch remaining (after map fit) outliers to the nearest rotamer
- restraints
- enabled�= True�Use rotamer restraints
- sigma�= Auto�The smaller the value, the stronger the restraints
- target�= max_distant min_distant exact_match fix_outliers�Choice for the reference conformation to restrain to
On Jun 10, 2021, at 3:20 PM, Oliver Clarke <[email protected]> wrote:
Hi,
Is there any way to turn off rotamer fitting/side chain optimization in phenix.real_space_refine? Sometimes it seems to do more harm than good, and repositions sidechains completely out of the density, but I couldn't find how to switch it off.
Cheers
Oli