IRIDA SISTR Plugin for Salmonella serotyping

This new IRIDA plugin allows deployment of SISTR, a tool for in silico Salmonella serotyping, in IRIDA platform. For faster and independent updates, SISTR built-in workflow were decided to be ported to plugin format. The analysis report page feature is only available since IRIDA version 20.09. In future SISTR releases we plan to generate reports also in csv and pdf formats.

Features

New to versions 1.1.1 and 1.1.1b

• Updated typing databases powered by SISTR v1.1.1
• Reporting of serotyping results via a pretty web-page (requires IRIDA >= 20.09)
  • More detailed analysis reports for MASH, cgMLST and BLAST profiles
  • Final results Quality Control checks
• Plugin versions 1.1.1 and 1.1.1b are similar with the later supporting IRIDA v21.01 new metadata update features

Building plugin

Compiled plugin version is available in *.jar folder, but you can build your own copy. Building and packaging this code is accomplished using Apache Maven. However, you will first need to install IRIDA to your local Maven repository. The version of IRIDA you install will have to correspond to the version found in the irida.version.compiletime property in the pom.xml file of this project. To build successfully plugin there is a need to compile IRIDA corresponding to the version specified in pom.xml. Here is a brief workflow to compile new *.jar file from the source code

IRIDA_VERSION=20.09
# Build IRIDA dependencies
git clone https://github.com/phac-nml/irida.git
git checkout ${IRIDA_VERSION}
#IRIDA dependencies will be located in ~/.m2
mvn install -DskipTests 
git clone https://github.com/phac-nml/irida-plugin-sistr.git
cd irida-plugin-sistr
# Build ECTyper plugin
mvn package -DskipTests  #find your package in /target
# move to *.jar to /etc/irida/plugins/

Below you will find more detailed explanations of each step above.

Install

As most IRIDA plugins, this plugin is readily installable by the placement of the *.jar file in /etc/irida/plugins directory. After IRIDA server restart, the new pipeline should appear in the list of pipelines.

Dependencies

The following dependencies are required to make and run this plugin.

• IRIDA >= 19.01.3 (analysis report rendering feature requires >= 20.09)
• Java >= 1.8 and Maven >= 3.3.9 (to build IRIDA dependencies)
• Galaxy >= 16.01
• Shovill == 1.1.0
• SISTR == 1.1.1

Galaxy configuration

The plugin assumes a properly configured Galaxy instance that will run the workflow included in the plugin. For this version of the plugin, the backend Galaxy instance needs to have the following tools installed via the Admin interface.

• Shovill v1.1.0

  • version 1.1.0
  • revision 6:83ead2be47b2
  • published 2020-07-03
  • ToolShed direct link

• SISTR v1.1.1

  • version 1.1.1
  • revision 4:17fcac7ddf54
  • published 2020-08-06
  • ToolShed direct link

Gallery

A couple of illustrations demonstrating plugin in action.

Feature updates

• inclusion of the following new fields in the IRIDA Project Line List output
  • cgmlst_found_loci
  • serovar_antigen
  • mash_serovar
  • mash_subspecies
  • cgmlst_genome_match
  • mash_genome
  • mash_distance
  • qc_messages
• update to the newest shovill version 1.1.0 with more relaxed assembly setting with minimum contig length of 1bp

Automatic metadata population

The plugin allows for automatic IRIDA project metadata population and automatic triggering upon sequencing data upload. These features need to be updated in the IRIDA project configuration page.

Figure 1: Pipeline Tile

Figure 2: Key metadata fields populated after serotyping run

Figure 3: Serotyping results rendered by `SISTR Viewer` into a pretty web-page

Troubleshooting

After successful installation, plugin should appear in the list of available pipelines. If not, check web-server log files for errors (e.g. Tomcat 7 logs /var/log/tomcat7/catalina.out)

Testing

Folder sampledata has

• stripped versions of the Salmonella Newport raw reads from SRR12168692 accession. These files could be used to test plugin in your IRIDA instance
• SISTR_results.json example results file output by the tool