You are here

Question about running scallo

9 posts / 0 new
Last post
Question about running scallo
#1

 I am trying to run code on https://github.com/strauchlab/scaffold_design

but it occur an error :

basic.io.database: [ WARNING ] Unable to locate database file scoring/score_functions/aa_composition/%%aa_comp%%.comp
Error: [ ERROR ] ERROR: Exception caught by JobDistributor while trying to get pose from job 'startingstub_0000_0008_0001'
Error: [ ERROR ] 

File: src/basic/database/open.cc:77
Database file open failed for: "scoring/score_functions/aa_composition/%%aa_comp%%.comp"

Error: [ ERROR ] 
Error: [ ERROR ] Treating failure as bad input; canceling similar jobs
protocols.jd2.FileSystemJobDistributor: job failed, reporting bad input; other jobs of same input will be canceled: startingstub_0000_0008_0001
protocols.jd2.JobDistributor: no more batches to process... 
protocols.jd2.JobDistributor: 1 jobs considered, 1 jobs attempted in 9 seconds
Error: [ ERROR ] Exception caught by rosetta_scripts application:

File: src/protocols/jd2/JobDistributor.cc:328
1 jobs failed; check output for error messages
Error: [ ERROR ] 
 

I have checked the /rosetta/main/database/scoring/score_functions/aa_composition fold, but only six files there without %aa_comp%.comp 

Thanks in advance for any responses on this matter

Category: 
Post Situation: 
Mon, 2021-11-29 23:25
danbo

I'm not familiar with that particular repository & protocol, but the double `%%` notation is for the "script_vars" facility in Rosetta. In the command line which actually launches the rosetta_scripts.linuxgccrelease executable, there should be an option `-parser:script_vars` which should be passed the parameter `aa_comp=filename`, where "filename" is a choice between one of the six options which are actually in that directory.

Since I'm not familiar with this protocol, I can't say for certain why that parameter isn't being set properly. It might be something you need to do (if you're building the command line yourself), or it might be a setting missing from one of the scripts which builds the command line for you. (e.g. you missed off a parameter when you ran the script.)

From a quick search of the repo, it looks like the parameter is set in the "generate_design_job.sh" script.

Tue, 2021-11-30 07:23
rmoretti

感谢#2 的回复。
我尝试使用选项“-parser:script_vars aa_comp=exactly_one_ala.comp”运行代码。它可以解决问题,但出现了另一个问题。
错误:无法解析指定的存储:/lab/databases/VALL_clustered/backbone_profiler_database_06032014
错误:: 退出:src/protocols/indexed_structure_store/FragmentStoreManager.cc 行:71
可能与之前的问题出于相同的原因。
如果每个人都可以帮助我,我将不胜感激

Thu, 2021-12-02 02:39
danbo

It's probably best if you keep the conversation here in English. 

Doing my best with Google translate, it sounds like the issue is that you don't have the /lab/databases/VALL_clustered/backbone_profiler_database_06032014 directory on your machine.

Was that something you specified? If so, you need to double check your paths and make sure that this directory is where you think it is.

If you didn't specify it yourself, this is something you'll need to change in the options files/scripts. You ned to figure out how to obtain the backbone_profiler_database, then download it to a directory on your own machine. Once you do that, edit the option files/scripts to point to the full path of that directory on your own machine.

Thu, 2021-12-02 07:34
rmoretti

Sorry for my carelessness, I will try to talk in English.

The problem is "Unable to resolve the specified storage: /lab/databases/VALL_clustered/backbone_profiler_database_06032014"

Maybe this means that when it tries creating a file under "/" (the root directory), it doesn't have permission to do so.

Please correct me if I am wrong

thanks

Thu, 2021-12-02 22:41
danbo

My guess is it's attempting to *read* from the specified directory. I believe that the "/lab/databases/VALL_clustered/backbone_profiler_database_06032014" directory is the location of that database on the computer of the person who originally put together the scripts you're using. Since you're in a different lab with a different computer system, you'll need to separately download the backbone_profiler_database_06032014 database. (Whatever that is -- check the documentation and the references for the protocol to see if it's mentioned.)

Once you have that downloaded, put it in a directory where you do have write access. That's unlikely to be under /lab. Since you've changed the location of the database, you'll need to edit the scripts and other files for the protocol, to replace the "/lab/databases/VALL_clustered/backbone_profiler_database_06032014" path with the actual path where you've placed the database.

Fri, 2021-12-03 06:41
rmoretti

I changed the location of the database to let it can locate the file "backbone_profiler_database_06032014" and it worked.

But it occured another question.(It really made me mad)

 

terminate called after throwing an instance of 'utility::json_spirit::Error_position'
Got some signal... It is:6
Process was aborted!
 

Please help me once more.

Thanks a lot

 

Wed, 2021-12-08 00:08
danbo

That error comes from not being able to properly read one of the input files. Without a further backtrace (or the associated ROSETTA_CRASH.log), I can't hazard a guess which file might be at issue, but I'd recommend double checking the backbone_profiler_database_06032014 directory, to make sure that things are set up appropriately.

Wed, 2021-12-08 10:24
rmoretti

I copy a backbone_profiler_database_06032014 from other sites from resetta, and it worked perheps.

Now it shows:

Error: [ ERROR ] Exception caught by rosetta_scripts application:

File: src/protocols/rosetta_scripts/ParsedProtocol.cc:401
Filter cavity not found in map

And it shows "core.select.residue_selector.util: [ WARNING ] Selector name is empty!" many times, I don't know if it matters.

Thanks for your help so many times 

 

Wed, 2021-12-22 23:51
danbo