NLRtracker: extracting NLRs from plant proteomes

26 November, 2021

Identification of NLR genes in annotated protein or transcript sequences

Introduction

NLRtracker extracts and functionally annotates NLRs from protein or transcript files based on the core features found in the RefPlantNLR dataset.

RefPlantNLR is a comprehensive collection of experimentally validated plant disease resistance proteins from the NLR family

Requirements & Installation

This is a pipeline to be run on unix based machines. The following software must be available in your path. At least 2G of free memory is needed for InterProScan but more memory is better.

InterProScan
- Requires Java 11
- If a different version than v5.53-87.0 is used specify the path to the description with -d
HMMER for functional annotation of the C-terminal jelly roll/Ig-like domain (C-JID). This only works for protein sequence input.
- Download v3.3.2 and make sure it is available in the environment
R version >= 4.1.0
R package
FIMO (MEME Suite version 5.2.0)

Using this script

Usage: NLRtracker.sh [OPTION]...
  -h               Display help
  (required)
  -s Filepath      File path to amino acid seqence file (.fasta)
  -o String        Directory name to save output

  (optional)
  -i Filepath      Result of Interproscan (.gff3)
  -f Filepath      Result of FIMO (.gff)
  -c Integer       Number of CPUs for interproscan
                   Default: 2
  -m Filepath      meme.xml for use with FIMO
                   Default: module/meme.xml (from NLR Annotator)
  -x Filepath      hmm for use with HMMER
                   Default: module/abe3069_Data_S1.hmm (from Ma et al., 2020)
  -d Filepath      Description of Interproscan
                   Default: module/InterProScan 5.51-85.0.list

Sample command

run NLRtracker in the same directory:

./NLRtracker -s sample_data/sample.fasta -o out_dir

if you already have results of interproscan and FIMO

bash NLRtracker.sh \
       -s sample_data/sample.fasta \
       -i sample_data/test_interpro.gff3 \
       -f sample_data/test_fimo.gff \
       -o test

Input & Option

(required)

-s … Amino acid sequence fasta
-o … Output directory name

(optional)

-i … Output of interproscan (interproscan.sh -i sample.fasta -appl Pfam,Gene3D,SUPERFAMILY,PRINTS,SMART,CDD,ProSiteProfiles -f gff3)
-f … Output of FIMO (fimo module/meme.xml sample.fasta)
-c … Number of CPUs to run interproscan. Default:2
-m … meme.xml file to run FIMO. Default:module/meme.xml
-x … hmm file to run hmmsearch Default:module/abe3069_Data_S1.hmm
-d … Description of Interproscan. Default: module/InterProScan 5.51-85.0.list

Output

Here is an overview of the output files created by the script. The files will be output by in the specified output directory and modified to include the output directory name.

The interpro, FIMO, and hmmer output
- fimo_out/
- interpro_result
- CJID.txt
NLR-associated (RPW8, TX, CCX, and MLKL genes):
- NLR-associated.lst: list with gene identifiers
- NLR_associated.fasta: fasta file
- NLR-associated.gff3: functional annotation of NLR-associated genes in gff3 format
The NLRs:
- NLR.lst: list with gene identifiers
- NLR.fasta: fasta file
- NLR.gff3: functional annotation of NLR genes in gff3 format
- NBARC.fasta: the extracted NB-ARC domains in fasta format
- NBARC_deduplictated.fasta: the extracted NB-ARC domains, identical sequences collapsed, in fasta format
Domain architecture of NLRtracker output for use with iTOL
- iTOL.txt
- iTOL_dedup.txt: for use with the deduplicated NB-ARC domains.
NLRtracker.tsv: the NLRtracker classification of each entry
Domains.tsv: The individual domains identified. This file can be used with the refplantnlR R package for drawing the NLR domain architecture

Versions

v1.0.3
- Update to InterProScan 5.53-87.0
- Improved fasta read-in (using Biostrings::readAAStringSet)
- Improved identification of C-JID domain (addition of PF20160 signature)
- Addition of extraction of sequences containing C-JID domain but lacking an NB-ARC domain
- Corrected a mistake fusing multiple sequential NB-ARC domains
- Corrected a mistake assigning CC_G10-NLRs as CC-NLRs in the column subclass (putative)
v1.0.2
- Addition of C-JID domain annotation using hmmer
- Addition of CblN domain (overlapping with RPW8-type CC domains)
- Addition of output for use with iTOL
v1.0.1
- Update to InterProScan 5.51-85.0
- Improved NB-ARC extraction (including winged-helix domain)
- Improved identification of Rx-type CC (addition of G3DSA:1.20.5.4130 signature)
v1.0.0
- NLRtracker

slt666666/NLRtracker