[phenixbb] Command line phenix.sculpt_ensemble

[Randy John Read]

Dear Matthew,

Sorry, I somehow missed this for a few days.

The first thing I would say is that I’m not sure there’s much of a use case for sculptor and ensembler (or the combined application you were using) in these days of AlphaFold. Sculpted ensembles were a really powerful way to combine information from a collection of distant homologues to make a better model for molecular replacement. However, since AlphaFold became available I think the current best approach in this scenario is to take the AlphaFold model (which will integrate everything known structurally about the class of proteins far better than ensembler), run it through process_predicted_model (to downweight or trim off the less reliable parts of the model) and use that single model as your molecular replacement model. I haven’t yet seen a counter-example where the experimental structures of the distant homologues are better. Of course, if someone has a counter-example that would be very interesting to learn about!

In terms of this bug, I’m wondering whether it’s a Python compatibility issue. I’m guessing that your Mac might be one of the newer ones with ARM (M1 or M2) chips. If so, are you using a build for the older Intel processors, or the native ARM build? In any case, if you’re using the ARM build then this will be built with Python 3. The easiest way to tell is to go to a terminal window with Phenix set up and type:

    phenix.python --version

This will report whether or not you’re using Python 2.7 or Python 3.7.

The sculpt_ensemble procedure has a lot of options, and I would need to know exactly which options you had set in order to try to reproduce the problem. Could you look in the project directory for a file with a name like .phenix/project_data/sculpt_ensemble_NN.eff, where “NN” is the job number? If you could send me that file, that would help.

Best wishes,

Randy Read

> From: Whitley, Matthew (NIH/NCI) [C] <matthew.whitley at nih.gov>
> Date: Thu, Sep 7, 2023 at 10:54 PM
> Subject: [phenixbb] Command line phenix.sculpt_ensemble
> To: phenixbb at phenix-online.org <phenixbb at phenix-online.org>
> 
> 
> Hello all,
>  I am trying to use the command line tool phenix.sculpt_ensemble to process a series of PDB files in preparation for molecular replacement.
>  However, when I run the following on the command line:
>  phenix.sculpt_ensemble 2*.pdb input.target_sequence=RRAS2.fasta
>  I get the following errors:
>  Traceback (most recent call last):
>   File "/Applications/phenix-1.21rc1-5015/build/../modules/phaser/phaser/command_line/sculpt_ensemble.py", line 84, in <module>
>     sculpt_ensemble.run( phil = params.phil, logger = logger )
>   File "/Applications/phenix-1.21rc1-5015/modules/phaser/phaser/sculpt_ensemble.py", line 610, in run
>     logger = logger,
>   File "/Applications/phenix-1.21rc1-5015/modules/phaser/phaser/sculpt_ensemble.py", line 259, in read_in_target_sequence_files
>     for ( fname, i ) in itertools.zip_longest( filenames, range( 1, nchains + 1 ) ):
> AttributeError: 'module' object has no attribute 'zip_longest'
>   I am running phenix 1.21rc1-5015-000 on Mac OS 13.5.1.
>  Any ideas?
>  Thanks,
> Matthew
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb
> Unsubscribe: phenixbb-leave at phenix-online.org

-----
Randy J. Read
Department of Haematology, University of Cambridge
Cambridge Institute for Medical Research     Tel: +44 1223 336500
The Keith Peters Building
Hills Road                                                       E-mail: rjr27 at cam.ac.uk
Cambridge CB2 0XY, U.K.                              www-structmed.cimr.cam.ac.uk