Hi all,
I'm running a software tool that requires a database path to NR database in the configuration yaml file, see snippet of yaml below:
---
max_threads: 4
# DB path
nr_db_path: [nr]
the issue I'm having is that I don't have the storage space to download the NR database on my computer. Is there any way to link the database without having to install it locally?
Thanks!
2 answers
You could use a network storage location but things will be slow as mud unless you have a high performance storage system and fast interconnects. So in practical terms the answer would likely be no.
The last time I used nr was more than 10 years ago despite doing daily protein searches. What I normally use is a UniProt database thinned down at 90% identity, which is less than 1/2 of nr in terms of size and a number of sequences. The reason is not to save on storage space - it is to save on search time because near-identical sequences are not needed to detect distant protein homologs. Your application may be different than mine.
Log in to answer this question.