You are here

ERROR: beta_nov16(.wts) requested, but -corrections::beta_nov16 not set to true. This leads to a garbage scorefunction.

7 posts / 0 new
Last post
ERROR: beta_nov16(.wts) requested, but -corrections::beta_nov16 not set to true. This leads to a garbage scorefunction.
#1

Hello everyone, while running a scoring file, I entered the following command line:

$/home/dengxj/software/rosetta_src_2021.16.61629_bundle/main/source/bin/rosetta_scripts.hdf5.linuxgccrelease -parser:protocol $CAO_2021_PROTOCOL/paper_monomer_design.xml -s HHH_bc_00001.pdb -script_vars runpsipred_single=/home/dengxj/software/psipred/runpsipred_single -indexed_structure_store:fragment_store /opt/software/ss_grouped_vall_all/ss_grouped_vall_all.h5 -parser:script_vars CAO_2021_PROTOCOL=$CAO_2021_PROTOCOL

However, an error message was obtained as follows:

basic.io.database: Database file opened: scoring/score_functions/EnvPairPotential/cenpack_log.txt
core.scoring.ramachandran: shapovalov_lib::shap_rama_smooth_level of 4( aka highest_smooth ) got activated.
basic.io.database: Database file opened: scoring/score_functions/rama/shapovalov/kappa25/all.ramaProb

ERROR: beta_nov16(.wts) requested, but -corrections::beta_nov16 not set to true. This leads to a garbage scorefunction.  Exiting.
ERROR:: Exit from: src/core/scoring/ScoreFunctionFactory.cc line: 225
Error: [ ERROR ] ERROR: Exception caught by JobDistributor while trying to get pose from job 'HHH_bc_00001_0001'
Error: [ ERROR ] Treating failure as bad input; canceling similar jobs

[ ERROR ]: Caught exception:


File: src/core/scoring/ScoreFunctionFactory.cc:225
[ ERROR ] UtilityExitException
ERROR: beta_nov16(.wts) requested, but -corrections::beta_nov16 not set to true. This leads to a garbage scorefunction.  Exiting.




AN INTERNAL ERROR HAS OCCURED. PLEASE SEE THE CONTENTS OF ROSETTA_CRASH.log FOR DETAILS.


protocols.jd2.FileSystemJobDistributor: job failed, reporting bad input; other jobs of same input will be canceled: HHH_bc_00001_0001
protocols.jd2.JobDistributor: no more batches to process... 
protocols.jd2.JobDistributor: 1 jobs considered, 1 jobs attempted in 8 seconds

---------------------------------------------------------------
[ ERROR ]: Error(s) were encountered when running jobs.
1 jobs failed;
Check the output further up for additional error messages.
---------------------------------------------------------------

 

The XML file(paper_monomer_design.txt)  used has been uploaded. So what may be the reason for the problem? Did I input the wrong command line? Or is there a problem with the scorefunction? I'm not sure. If you have any ideas, please provide me with some suggestions. Thank you.

AttachmentSize
paper_monomer_design.txt11.76 KB
Category: 
Post Situation: 
Mon, 2023-05-08 06:36
sia

In your XML, you're calling for the beta_nov16 scorefunction weights. (Not too unusual - the Baker lab uses them frequently.)

However, to properly use beta_nov16, you also have to change the scoring environment (the command line options). The current default scoring environment is REF2015, which isn't quite compatible with the beta_nov16 weights. (Well, its compatible in the sense things might run, but it won't give you the results that you expect from that scorefunction.) Adding `-corrections::beta_nov16` to the command line should be sufficient to switch environments.

Mon, 2023-05-08 07:32
rmoretti

Hey rmoretti, Thank you very much, your suggestion is very helpful to me, and it has been successfully run now.

Mon, 2023-05-08 18:56
sia

I have successfully calculated the score of protein structure with paper_monomer_design.xml(which has been uploaded earlier.), but the calculate process takes a little long, a structure with more than 50 amino acids takes about half an hour, and the output scoring file contains many items. I only need a few scores such as mismatch_probability 、percent_core_SCN、score_per_res、ss_sc、worst9mer and worst9mer_helix, so I want to change paper_monomer_design.xml. Maybe I can make the scoring process faster by deleting some unnecessary scoring items. But I'm not sure which items are necessary for the calculation process. Can anyone give me some advice? Can it be deleted and how should it be deleted?Or are there other ways to make the calculation process faster? Thankyou!

The output scoring file has been uploaded.

File attachments: 
Sun, 2023-05-14 19:38
sia

The reason thinks take a bit on the long side is the fact that you're running FastDesign 5 times. You can monitor the log file, but I'm guessing that the majority of the time is spent in the design phase, rather than the scoring phase.

That said, if you don't need score terms, the way to remove them is by editing the PROTOCOLS block. All fo the filters listed there contribute a score term to the scorefile. So fi you don't need AlaCount, you can remove the PROTOCOLS line which lists AlaCount.

Note, though, that a number of the terms in the file (e.g. fa_atr, omega, ref, hbond_sc, etc) are standard Rosetta scoreterms which get output for pretty much any structure, even without separate calculation. But I shouldn't worry about them, as they're (relatively) quick to compute and output, and should contribute less than a second per output structure.

Mon, 2023-05-15 07:16
rmoretti

So can I remove 5 times fastdesign? Does it have an impact on the scoring results?

Mon, 2023-05-15 18:28
sia

FastDesign is the design step. It has an effect on scoring by actually changing the protein structure and sequence.

If all you want to do is rescore your inputs, then yes, you *definitely* want to remove the FastDesign steps. You also want to remove the PackRotamersMover step. You would also need to carefully consider any other movers which might change the structure. (I think the StructProfileMover only changes Pose annotations, and doesn't affect the coordinates of the structure.)

If you're actually interested in doing the design, then you don't want to remove the FastDesign steps, as that would remove the design. You might be able to play around with reducing the number of FastDesign steps, as five might be overkill for your situation, but you may have to benchmark it and test various levels out to see what the difference in design quality is.

Tue, 2023-05-16 07:10
rmoretti