Update __init__.py#528
Conversation
Update ensemble_refinement to 2020 version. Added DEN restraints (now used by default). Fixed refinement flag manager. Added support for importing external TLS parameters from additional PDB.
|
Tom
Can you tell me why DEN is default?
Do you have any tests for these changes?
Cheers
Nigel
…---
Nigel W. Moriarty
Building 33R0349, Molecular Biophysics and Integrated Bioimaging
Lawrence Berkeley National Laboratory
Berkeley, CA 94720-8235
Phone : 510-486-5709 Email : NWMoriarty@LBL.gov
Fax : 510-486-5909 Web : CCI.LBL.gov
On Wed, Sep 16, 2020 at 9:09 AM Tom Burnley ***@***.***> wrote:
@tomburnley <https://github.com/tomburnley> requested your review on: #528
<#528> Update *init*.py.
—
You are receiving this because your review was requested.
Reply to this email directly, view it on GitHub
<#528 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAHYQPDIXHD5S4EZ74YCT3DSGDPLDANCNFSM4RPB7OWQ>
.
|
|
Hi Nigel,
DEN -> we found in testing that it makes the simulations more stable
without negatively affecting sampling. W/out DEN on some occasions parts
of the structure can unfold out of density. The attractive terms in Amber
may alleviate this but for now w/ the repel potential it's better. The DEN
pairs are regenerated multiple times during the simulation so sampling does
not seem to be compromised.
Testing -> yes... see separate email.
Cheers,
Tom
On Wed, 16 Sep 2020 at 17:15, Nigel W. Moriarty <notifications@github.com>
wrote:
… Tom
Can you tell me why DEN is default?
Do you have any tests for these changes?
Cheers
Nigel
---
Nigel W. Moriarty
Building 33R0349, Molecular Biophysics and Integrated Bioimaging
Lawrence Berkeley National Laboratory
Berkeley, CA 94720-8235
Phone : 510-486-5709 Email : ***@***.***
Fax : 510-486-5909 Web : CCI.LBL.gov
On Wed, Sep 16, 2020 at 9:09 AM Tom Burnley ***@***.***>
wrote:
> @tomburnley <https://github.com/tomburnley> requested your review on:
#528
> <#528> Update *init*.py.
>
> —
> You are receiving this because your review was requested.
> Reply to this email directly, view it on GitHub
> <#528 (comment)>, or
> unsubscribe
> <
https://github.com/notifications/unsubscribe-auth/AAHYQPDIXHD5S4EZ74YCT3DSGDPLDANCNFSM4RPB7OWQ
>
> .
>
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#528 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACSF77V6VBGGQ2ZLVTQSFHLSGDQBXANCNFSM4RPB7OWQ>
.
|
olegsobolev
left a comment
There was a problem hiding this comment.
Just a note, there are bunch of functions of the model class that deal with TLS:
get_header_tls_selections()
extract_tls_selections_from_input()
determine_tls_groups(self, selection_strings, generate_tlsos)
I see a lot of code to deal with TLS in this particular file, so maybe something more is going on, so I would not insist on using them in this particular case instead of things like
self.model.tls_groups.tlsos = fit_tlsos
fit_tlsos = [tls_tools.tlso(t=o.t, l=o.l, s=o.s, origin=o.origin) for o in tls_params.tls_params]
| make_header("Import External TLS", out = self.log) | ||
| print('External TLS model: ' + self.params.import_tls_pdb, file=self.log) | ||
| pdb_import_tls = self.params.import_tls_pdb | ||
| pdb_tls_inp = iotbx.pdb.hierarchy.input(file_name=pdb_import_tls) |
There was a problem hiding this comment.
please use iotbx.pdb.input() instead, preferably checking if the file exists.
There was a problem hiding this comment.
Still this, with the explanation in the previous message.
Move all import statements to header. Remove duplicate self.assign_solvent_tls_groups() in Import TLS from reference model.
|
There are unused imports causing the Checks tests to fail. An update will cause the CI to run again and the error for Full and Base should go away since I checked in some updates to the |
Remove unused imports
|
Thanks! The Ci branch failures are fine. That pipeline does not run on the final, merged code with the updates to |
|
This requested change seems to be missed:
|
Can you explain the need for this and point to an example of use I can copy from please. |
|
@tomburnley , iotbx.pdb.hierarchy.input() seems to be a very thin wrapper that pretty much constructs hierarchy and keep input object: cctbx_project/iotbx/pdb/hierarchy.py Lines 2493 to 2536 in 1be58fd The original reason to create an extra layer for such a simple task is not clear to me in the first place. I could very well miss something, so clarifications from anyone are welcomed. As of now, I believe this whole layer better be removed, therefore advocate against using it anywhere. In practical terms regarding this pull request I suggest to change this: to this: Other examples of iotbx.pdb.input() usage can be found e.g. in iotbx/command_line/pdb_labels_comparison.py and many other files across the repository. |
Updated version of ensemble refinement to include den restraints and echt b-factor model to accompany paper submitted to Acta D. Tested with phenix-1.19.2-4158. Issues with previous pull request (#528 (comment)) fixed.
Updated version of ensemble refinement to include den restraints and echt b-factor model to accompany paper submitted to Acta D. Tested with phenix-1.19.2-4158. Issues with previous pull request (#528 (comment)) fixed.
Updated version of ensemble refinement to include den restraints and echt b-factor model to accompany paper submitted to Acta D. Tested with phenix-1.19.2-4158. Issues with previous pull request (cctbx#528 (comment)) fixed.
Update ensemble_refinement to 2020 version. Added DEN restraints (now used by default). Fixed refinement flag manager. Added support for importing external TLS parameters from additional PDB.