Qiime/Qiime2 on the Hoffman2 cluster

A centralized installation of qiime2 is available via anaconda:

To run:

qrsh 
module load anaconda

source /u/local/apps/anaconda2/etc/profile.d/conda.csh

conda info -e

conda env list

and load the needed environment. For example for qiime2-2019.4 you can do so with:

source activate qiime2-2019.4

you can now test that things are working with:

qiime --help

to exit from the conda environment please use:

source deactivate

To use qiime2 from a batch job you will need to create a script, called for example submit_qiime.sh, that has these commands in it:


##### SUBMIT QIIME2 SCRIPT START #######
#!/bin/bash
#$ -cwd
# error = Merged with joblog
#$ -o joblog.$JOB_ID
#$ -j y
# RESOURCE REQUESTED:
#$ -l h_rt=1:00:00,h_data=1G
#$ -pe shared 1
# Email address to notify
#$ -M $USER@mail
# Notify when
#$ -m bea

# echo job info on joblog:
echo "Job $JOB_ID started on: " `hostname -s`
echo "Job $JOB_ID started on: " `date `
echo " "

# load the job environment:
. /u/local/Modules/default/init/modules.sh
source activate qiime2-2019.4


# run qiime2 code, for example:
echo 'qiime --help'
qiime --help

source deactivate

# echo job info on joblog:
echo "Job $JOB_ID ended on: " `hostname -s`
echo "Job $JOB_ID ended on: " `date `
echo " "
##### SUBMIT QIIME2 SCRIPT END #######



Please change:

# run qiime2 code, for example:

echo 'qiime --help'

qiime --help

to reflect the actual qiime commands you intend to use.

Also please change:

# RESOURCE REQUESTED:

#$ -l h_rt=1:00:00,h_data=1G

#$ -pe shared 1

to reflect the time (h_rt), memory (h_data) and number of cores (shared X) you intend to use (you may also need to add to the "#$ -l" line the complex "highp" if you intend to run your code on your group resources (you would then use: "#$ -l h_rt=HHH:00:00,h_data=XG,highp").

When you are done editing the submission script issue:

chmod u+x submit_qiime2.sh

and you can then submit your job with:

qsub submit_qiime2.sh

__
RD - October, 2017
RD - Last updated June 2019