DRAM
DRAM copied to clipboard
WrightonLabCSU
subprocess.CalledProcessError
Hello, We are facing error: "Failed to mmap memory dataSize=80676316565 File=DRAM_data/database_files/pfam.mmsmsa. Error 12." while preparing for database.
It is halting with "subprocess.CalledProcessError: Command '['mmseqs', 'msa2profile', 'DRAM_data/database_files/pfam.mmsmsa', 'DRAM_data/database_files/pfam.mmspro', '--match-mode', '1', '--threads', '10']' returned non-zero exit status 1."
Any suggestions to resolve it?
Hi!
How much RAM is available on your system? It looks like there might not be enough RAM assigned to run this step.
Mike
Hello!
Currently, we are using a system with 16GB. What is the minimum RAM requirement while skipping uniref90 and commercial KEGG?
Parth
Unfortunately, even with skipping those databases, you may still need ~64GB for this step. You still have 2 options.
- You can try using dram through kbase.
- @shafferm has uploaded a minimum database set to zenodo, you can download the files and use
DRAM-setup.py set_database_locationsto skip this memory intensive step. However, there are problems with this approach, as detailed in this issue thread.
Good luck.
@rmFlynn Thank you. The second option worked for us. We are using DRAMv with the output of virsorter2, but it is giving an error given below:
ValueError: No virsorter calls found in for-dramv/viral-affi-contigs-for-dramv.tab for scaffold V1_1_(paired)_contig_51-cat_1 from input fasta.
If the viral-affi-contigs-for-dramv.tab file is skipped during annotation then it gives an auxiliary score error for the distill module. Is there any way to solve this issue?
In this case the problem could be the "(", ")" characters. If you have not already, run DRAM-v.py remove_bad_characters -v tabfile.tab and clean_tabfile.tab and DRAM-v.py remove_bad_characters -i input_fasta.fa clean_input_fasta.fa this may not remove the parens, so you may need to run your own command replacing "(" and ")" with "_". If running with the cleaned output does not work, then search for any matches in the viral-affi-contigs-for-dramv.tab that have the parens replaced, for example grep "V1_1_.*paired.*_contig_51-cat_1" viral-affi-contigs-for-dramv.tab. If there are no matches, then we need to figure out where that name went.