Nextflow InterProScan Production Manager (NF-IPM)

A Nextflow pipeline to coordinate the calculation, import and cleaning of InterProScan matches for the InterPro database.

Getting started

Requirements:

• Nextflow >= 25.04.6

Installation:

git clone https://github.com/ProteinsWebTeam/iprscan-manager.git

Configuration

The IPM pipeline relies on two configuration file. Templates can be found in the conf directory.

General configuration

The conf/ipm.conf is used for the general configuration of IPM and InterProScan.

Warning

The InterProScan6 work directory will be extremely large! Make sure to point the interproscan.workdir field to a suitable location.

• databases - configure database connections
  • intprscan: configuration for the postgre-sql interproscan database
    • uri (//Host:Port/Service)
    • user
    • password
    • engine: postgresql
  • uniparc: configuration for the UATST uniparc database (the read-only database)
    • uri (@Host:Port/Service)
    • user
    • password
    • engine: oracle
• interproscan - configure how InterProScan6 is run
  • device - specify the configuration for cpu or gpu execution
    • executable: "ebi-pf-team/interproscan6" or path to a local InterProScan6 installation main.nf file
    • executor: Run InterProScan6 locally (local) or on SLURM (slurm)
    • container: Container runtime to use (e.g. 'docker' or 'baremetal' when the latter is supported)
    • profile: Specify any other InterProScan profiles to be used, e.g. bulk
    • workdir: Path to build the workdir. This directory can become extremely large!
    • maxWorkers: Set the --maxWorkers option in interproscan6
    • config: [Optional] Path to an Iprscan 6 config file. This needs to be used when running liscened software, otherwise iprscan will not know where to find the SignalP, Phobius and DeepTMHMM databases

Application resource configuration

The resources allocated to instance of InterProScan is dependent on the application (or member database) that is running. These values are defined in the conf/applications.conf file.

The applications configuration is included by default in IPM so you only need to the appsConfig configuration in your configuration file if you are deferring from the values in this file.

• appsConfig applications All applications default to the light configuration. To specify a different resource configuration list the application here * <application-name> = "<label>"
  • resources Define the resources for each label
    • <label> { memory = "X.GB", time = "Y.h" }

Usage

There are two required arguments:

Two arguments are required:

1. -c - Path to the imp.conf file
2. --methods - The name of the subworkflows (case-insensitive) as a comma separated list to run:
   • import
   • analyse
   • clean

To see all optional arguments use the --help option. Note some optional arguments are subworkflow specific.

Import

The IMPORT subworkflow coordinates retrieving all (or the latest) protein sequences from the UniParc database and importing them into the InterProScan (e.g. ISPRO) database.

There are 2 optional arguments:

1. --top-up - Import new sequences only
2. --max-upi - Maximum sequence UPI to import

For example:

nextflow run main.nf -c conf/imp.conf --methods import --top-up

Analyse

The ANALYSE subworkflow coordinates running InterProScan for every "active" analysis in the ISPRO.ANALYSIS table, and persists all results in the ISPRO database.

There is one optional argument:

1. --batch-size - The maximum number of sequences to be analysed by each instance of InterProScan 6

For example:

nextflow run main.nf -c conf/imp.conf --methods analyse

Clean

The CLEAN subworkflow deletes obsolete data for analyses listed as active 'Y' in the IprScan database.

There is one optional argument:

1. --analyses - IDs of analyses to clean (default: all)

For example:

nextflow run main.nf -c conf/imp.conf --methods clean