brinkmanlab/psortb_commandline repository overview

⁠PSORTb

PSORTb is a bioinformatics tool for predicting subcellular localization for a given set of protein sequences. The protein sequences must belong to one type of organisms, classified by cell membrane type in order to more accurately predict subcellular localization. The supported organism types are:

• Gram negative (-n)
• Gram positive (-p)
• archaea (-a)
• Gram negative without outer membrane
• Gram positive with outer membrane

The PSORTb program has been added to a Docker container to simply the installation process. Please scroll further down this page to find an Apptainer/Singularity workflow for this Docker image.

This PSORTb installation uses PSORTb version 3.  

⁠Docker Workflow

⁠Installing the Docker image

Warning: Users of this PSORTb app must have sudo access to run docker commands. Either that or an administrator must give you access to run "docker run" commands. If you do not have sudo access on the cluster where PSORTb will be run, consider using the Apptainer workflow instead (details further down this page).

This image takes up 2.55GB space.

% sudo docker pull brinkmanlab/psortb_commandline:1.0.2

Once the Docker image has been installed, the following script can be installed to simplify the running of PSORTb analyses:

% wget https://raw.githubusercontent.com/brinkmanlab/psortb_commandline_docker/master/psortb
% chmod +x psortb

Warning: Users of this PSORTb app must have sudo access to run docker commands. Either that or an administator must give you access to run "docker run" commands. If you do not have sudo access on the cluster where PSORTb will be run, consider using the Apptainer workflow instead (details further down this page).  

⁠Running a PSORTb analysis using the Docker image

Once the Docker image has been installed, run an PSORTb analysis like this:

% ./psortb -h
% ./psortb -i <FASTA format sequence file> -r <local results directory> [--positive|--negative|--archaea]

⁠Github repository

For administrators wanting to tweak the Docker setup, the Dockerfile used to build this image can be found on GitHub: https://github.com/brinkmanlab/psortb_commandline_docker⁠ The Dockerfile builds the complex environment of PSORTb. The repository also contains a wrapper script (named psortb) to run the PSORTb analysis inside the Docker container (i.e a wrapper for the "docker run" command).

Warning: Users of this app must have sudo access to run docker commands. Either that or an administrator must give you access to run "docker run" commands. If you do not have sudo access on the cluster where PSORTb will be run, consider using the Apptainer workflow instead (details further down this page).  

⁠Apptainer Workflow

Because Docker is often not available on clusters, we provide a way to use Apptainer instead, to give the same results. Apptainer can be run without sudo access although the initial build does require sudo access. In this case, it is possible to create the Apptainer image file (sif file) on a VM/system where you have root permissions and then copy the sif file into the system you plan to run your analyses.

⁠Apptainer Requirements

You need Apptainer 1.1.3 or newer available in your environment. In many academic clusters, this can be easily enabled by running a module.

Secondly, you will need to build an Apptainer image file (a sif file) from the PSORTb command line Docker image. Note building the sif file needs to be run as the root user. Because of this, you may need to create the sif file on a different VM and then copy it over to your VM of choice. Create the sif file like this:

% sudo apptainer build psortb.sif docker://docker.io/brinkmanlab/psortb_commandline:1.0.2

  Note, you can name this sif file anything you like, but psortb.sif is the default name used by the psortb_app wrapper script. If you name it anything else, you will need to use the --sif-file <myfile.sif> option when running the psortb_app wrapper.

⁠The psortb_app wrapper script (using Apptainer)

The psortb_app wrapper interacts with the Apptainer container for you so you only need to run a single command to submit an analysis. This wrapper script has lots of defaults set for your convenience, but these can be changed using available options. Download this script by running:

% wget https://raw.githubusercontent.com/brinkmanlab/psortb_commandline_docker/master/psortb_app

  Then make this file executable by running:

% chmod +x psortb_app

  View the options by running:

% ./psortb_app --help

  At minimum you will need the following options:

./psortb_app -i <fasta_sequence_file> -r <directory_to_put_results_file_in> -n|-p|-a

For example, for a Gram negative analysis:

% ./psortb_app -i myseq.txt -r results/ -n

  The -n, -p or -a option is required. It lets you specify whether your results set is Gram positive (-p), Gram negative (-n) or Archaea (-a). The input sequence file for each PSORTb analysis should contain only one type of bacteria or archaea. This allows much more accurate results.

⁠Versions

Jun 2017: Tag 1.0.0 - the first PSORTb command line Docker image release. Nov 2017: Tag 1.0.1 contains an additional 32 sequences in the subcellular localization blast data sets (see the PSORTdb paper⁠ for details). Jun 2018: Tag 1.0.2 contains a bug fix which addresses a change in Bioperl and ensures input sequences entered in lower case give the same results to those entered in upper case.

⁠License

PSORTb is distributed under GNU General Public License Version 3⁠.  

⁠Support

Please email any questions to [email protected]⁠

This application was developed by Gemma Hoad of the Brinkman Laboratory at Simon Fraser University, Greater Vancouver, Canada.