bioinformatics-centre / kaiju

Fast taxonomic classification of metagenomic sequencing reads using a protein reference database
http://kaiju.binf.ku.dk
GNU General Public License v3.0
264 stars 68 forks source link

Failure to creating the reference database and index of nr_euk #233

Open AlexSongh opened 2 years ago

AlexSongh commented 2 years ago

Hi I have a problem when downloading the euk_nr database using the slurm scripts I have installed a kaiju conda environment, and run the following codes

My HPC settings were

SBATCH --ntasks-per-node=8

SBATCH --mem-per-cpu=16GB

SBATCH --nodes=1

SBATCH --cpus-per-task=1

my codes are as follows:

source ~/.bachrc 
conda activate kaiju 
cd /nfs/turbo/cee-raskin/hangsong
mkdir kaijudb
cd kaijudb
kaiju-makedb -s nr_euk

I have following issues, do you know what's wrong and how to solve this?

image

Thanks!

pmenzel commented 2 years ago

Hm I don't know. Maybe the file got corrupted during download.

btw, check the README to see how much memory you need for building the database.

AlexSongh commented 2 years ago

I have enough memory in the storage account. Which I have 3 TB storage. But when I download it, it showed broken pipe. Then I changed to a slurm scripts. Do you know how to solve this issue?

image
pmenzel commented 2 years ago

I was thinking about RAM, not storage. Maybe it's better for you to download the index from the web server, see https://kaiju.binf.ku.dk/server

AlexSongh commented 2 years ago

Thank you!