Softwares¶

Is a software is already installed on the cluster?¶

Find a software¶

On our Web site: go to Resources -> Softwares

On front-end servers

with command search_module (case insensitive):

search_module <soft-name|part-of-soft-name>

where you replace <soft-name|part-of-soft-name> by (a part of) the soft name.

Example

search_module blastbioinfo/blast-2.2.26
bioinfo/ncbi-blast-2.2.29+
bioinfo/ncbi-blast-2.6.0+
bioinfo/RMBlast-2.6.0
bioinfo/samblaster-v.0.1.24
bioinfo/WuBlast2.0

with ls command:

ls /usr/local/bioinfo/src|grep -i <soft-name|part-of-soft-name>

where you replace <soft-name|part-of-soft-name> by (a part of) the soft name

with module command:

# Display all available software installed on the clustermodule avail# Search for module matching a (part of the) software namemodule search <soft-name|part-of-soft-name># Display available versions for a specific application# (with category in bioinfo,compiler,mpi or system) (case sensitive)module avail <category>/<soft-name># Display available versions for a specific applicationmodule avail -t 2>&1 | grep -i <soft-name|part-of-soft-name>

where you replace <soft-name|part-of-soft-name> by (a part of) the soft name, <soft-name> by the soft name and <category> by bioinfo, compiler, mpi or system

Examples

module avail bioinfo/cutadaptbioinfo/cutadapt-1.14-python-2.7.2 bioinfo/cutadapt-1.14-python-3.4.3
module avail -t 2>&1 | grep -i openmpimpi/openmpi-1.6.5
mpi/openmpi-1.8.1
mpi/openmpi-1.8.8
mpi/openmpi-1.8.8-intel2018.0.128
mpi/openmpi-2.1.2
mpi/openmpi-2.1.2-intel2018.0.128

Find a python module¶

On front-end servers with alias command search_Python_package (case insentitive):

search_Python_package <module-name>

Find an R package¶

On front-end servers with alias command search_R_package (case sensitive):

search_R_package <package-name>

Where are installed software on cluster?¶

Bioinfo: /usr/local/bioinfo/src
Compilers: /tools/compilers
Libraries: /tools/lib
Languages (Python, R , Java, Singularity..):
- /tools/statistics,
- /tools/containers,
- /tools/devel
Other system tools : /tools/others_tools
Useful scripts: /tools/bin (sarray, squota_cpu, saccount_info, ...). In user’s default PATH.

Ask for a software installation on cluster¶

Software installation or update are performed by complete the following form: Software installation

Run a software¶

Run a specific application

We Environment Modules to dynamically set up environments for different applications (see "How to use module command ?" section of this FAQ for more information).

To run a software you need to load the corresponding module.

Examples

Blast

# We load blastmodule load bioinfo/blast-2.2.26# We check that blast is availablewhich blastall/usr/local/bioinfo/src/NCBI_Blast/blast-2.2.26/bin/blastall

Python 2.7.2

# We load Python modulemodule load system/Python-2.7.2# We check that python is availablewhich python/tools/python/2.7.2/bin/python

Help to run a software¶

Software documentation¶

Typically, documentation is provided by the software in the installation folder (/usr/local/bioinfo/src/<soft-name>) and on the website of the software (we give the link to the website of each software on the Software page).

Use on SLURM cluster¶

You can also find specific information in a How_to_use_SLURM_<soft-name> file in the software installation directory (/usr/local/bioinfo/src/<soft-name>) or on our web site Software page (Avaibility/Use column, click on SLURM cluster link) .

Example directory¶

In the software installation directory, there is often a basic example submission job in an example_on_cluster directory (you need to copy it in your work space to run it to test, otherwise you'll have permission issues).

Example with bowtie

ls -1 /usr/local/bioinfo/src/bowtie/bowtie-1.2.1.1
bowtie-1.2.1.1-linux-x86_64.zip
bowtie2-2.2.9
bowtie2-2.3.3.1
bowtie2-2.3.3.1-linux-x86_64.zip
example_on_cluster
How_to_use_SLURM_bowtie
ls -1 /usr/local/bioinfo/src/bowtie/example_on_cluster/error.txt
example
lambda_virus.1.bt2
lambda_virus.2.bt2
lambda_virus.3.bt2
lambda_virus.4.bt2
lambda_virus.rev.1.bt2
lambda_virus.rev.2.bt2
output.txt
test_v2-2.2.9.sh

Others useful commands¶

On front-end servers, to find the path of the How_to_use_SLURM_<soft-name> file, use module help command:

module help bioinfo/bowtie2-2.2.9----------- Module Specific Help for 'bioinfo/bowtie2-2.2.9' -----------

See How_to_use file: /usr/local/bioinfo/src/bowtie/How_to_use_SLURM_bowtie

In your web browser, to view all How_to_use_SLURM_<soft-name> files: https://web-genobioinfo.toulouse.inrae.fr/How_to_Softs/

How to use module command ?¶

The basic command to use is module

# Print usage instructionsmodule# List available software modulesmodule avail# Add a module to your environmentmodule load <module-name># Unload/remove a modulemodule unload <module-name># Unload/remove all modulesmodule purge# Show what changes a module will make to your environmentmodule show <module-name> # Show the path to the 'How_to_use_SLURM_<soft-name>' filemodule help <module-name>

For more documentation, you can see the Environment Module website.

How to use R on the cluster ?¶

Tutorial describing how to run R scripts and compile Rmarkdown files on the cluster is available in tutorial part

How to find an already installed R package?¶

On front-end servers with alias command search_R_package (case sensitive):

search_R_package <package-name>

How to find if a Python package is already installed?¶

On front-end servers with alias command search_Python_package (case insentitive):

search_Python_package <module-name>

Java¶

Launch java onto the nodes for the java command:

java -Xmx4g

Choose a specific java version:

# We search available java versionsearch_module javadevel/java/17.0.6
devel/java/1.8.0_391
# We load javamodule load devel/java/17.0.6

Available compilers¶

To see available compilers:

module avail compiler

Exemple to load a compiler:

module load compilers/gcc/7.2.0

Create your own module¶

Documentation is available on the Environment Module website

A module file follows this template:

modulefile

#%Module1.0################################################################ (1)

module load <existant-module> ;#(2)!

prepend-path PATH </absolute/path/to/my/software/bin/directory/>

setenv <MY_ENV_VAR> <my-env-var-value> ;# (3)!

Mandatory header
if needed
if needed

It can be loaded the following way

module load -f </path/to/my/modulefile>

Examples: MultAlin

We define a module file for multalin software in such a way:

~/save/modules/multalin-5.4.1

#%Module1.0###############################################################

module load bioinfo/bowtie2-2.2.9
module load bioinfo/bwa-0.7.15
module load bioinfo/samtools-1.4

prepend-path PATH /usr/local/bioinfo/src/MultAlin/multalin.5.4.1/
setenv MULTALIN /usr/local/bioinfo/src/MultAlin/multalin.5.4.1/

Then we load the module file:

module load -f ~/save/modules/multalin-5.4.1# We check that related dependencies are also loaded:# - bioinfo/bowtie2-2.2.9which bowtie2/usr/local/bioinfo/src/bowtie/bowtie2-2.2.9/bowtie2
# - bioinfo/bwa-0.7.15which bwa/usr/local/bioinfo/src/bwa/bwa-0.7.15/bwa
# - bioinfo/samtools-1.4which samtools/usr/local/bioinfo/src/samtools/samtools-1.4/samtools

Dependencies can be unloaded individually:

module unload bioinfo/bowtie2-2.2.9

Your module can be unloaded the usual way:

module listCurrently Loaded Modulefiles:
1) bioinfo/bowtie2-2.2.9 2) bioinfo/bwa-0.7.15 3) bioinfo/samtools-1.4
4) /home/<username>/save/modules/multalign-5.4.1
module unload /home/<username>/save/modules/multalign-5.4.1# Alternative way to unloadmodule unload -f ~/save/modules/multalin-5.4.1

and all modules can also be unloaded:

module purge

Default directory for some software and `/home` overquota¶

By default, many softwares like .nextflow, .conda, .beast, .ncbi or .R store data in the home directory, which is limited by a small quota are usually located at your home directory whose allocated space is very limited.

A simple solution consist in creating (or moving if it already exist) this directory in your work space and then creating into your home a symbolic link to this directory.

Example with nextflow:

mkdir ~/work/.nextflow # or mv ~/.nextflow ~/work/ln -s ~/work/.nextflow ~/.nextflow

Example with conda:

mkdir ~/work/.conda # or mv ~/.conda ~/work/ln -s ~/work/.conda ~/.conda