# RNA update library

3 posts / 0 new
RNA update library
#1

Hi,

I'm building RNA structures of 4kb using rna_denovo. For some of my structures I have this error message:

ERROR: hey update AtomLevelDomainMap & RNA_ChunkLibrary to not use magic numbers like 999 and 1000

ERROR:: Exit from: src/core/import_pose/libraries/RNA_ChunkLibrary.cc line: 551

protocols.jd2.JobDistributor: [ ERROR ]

[ERROR] Exception caught by JobDistributor for job S_000001

[ ERROR ]: Caught exception:

File: src/core/import_pose/libraries/RNA_ChunkLibrary.cc:551

[ ERROR ] UtilityExitException

ERROR: hey update AtomLevelDomainMap & RNA_ChunkLibrary to not use magic numbers like 999 and 1000

The command line I'm running is:

rna_denovo.default.linuxgccrelease -fasta file.fa -secstruct_file file_struct -out:file:silent output_file.out -minimize_rna

Any idea of what could be causing this issue and how to solve it?

Many thanks :-)

Category:
Post Situation:
Mon, 2022-05-16 16:04
Anne Klein

It looks like the issue is that you're doing a modeling job that's large enough that you're running into a limit that the programmers didn't think rna_denovo would handle. It does look like this could be fixed in the code, but that won't really help you at the moment.

One option is to reduce the size of the modeling job. I think that something under 1000 nt or so would be safe, but you may have to play around a bit to see what triggers it.

The other option would be to set -rna::denovo::bps_moves false on the command line. The big drawback of this is that this turns off one of the modeling moves in the rna_denovo protocol, so the structures you get out of it may or may not be any good. I'm not all that familiar with the details of the protocol, but that might be reasonable -- it does look like its somewhat reasonable to do, as the -new_fold_tree_initializer true option required bps_moves to be false, and the DRRAFTER protocol makes use of that. Although it looks like using the -rna::denovo::secstruct_general option means that bps_moves becomes obligate.

Tue, 2022-05-17 08:06
rmoretti

Thanks! Adding  "-rna::denovo::bps_moves false" does the job for my simulations.

Tue, 2022-05-24 21:38
Anne Klein