Repository files navigation

NLRextract was written to search for NLR related hmms in multi-fasta proteinsequences. It uses hmms from pfam v32. It includes the search for the CC domain related MADA motif [1]. This tool is under development so let me know if something's odd or can be improved. I tested the performance of NLRextract againstNLRparser[2] andNLRtracker[3] in a small article on my website:https://www.biotinkertech.eu/project_NLRextract.html

IMPORTANT!!
24.02.22 - Potentially wrong output of NLRextract using system supplied awk:

Filtering of hmmrsearch / hmmrscan output was so far done with the system supplied awk version. Standard Installation of Ubuntu 20.04 comes with mawk as awk version. This version of awk has problems with scientific notation of numbers. Therefore, gawk=5.1.0 was included in environment.yml and the bash script was 'NLRextract' was adjusted to use gawk for filtering of results. If you have run NLRextract on standard Ubuntu 20.04 please make sure to rerun your analysis with the updated version!!!

IMPORTANT!!

• Searching for: NB-ARC, CC, TIR, RPW8 and LRR domains in the proteins
• Create a venn diagram from the domain infromation
• Extract the overlaps of the different domain combinations
• Extract sequences of proteins with the different domain combinations
• Plot number of different domains, and number of different NLRs

Please have a look at the environment.yml file for more information

(base) 💻 daniel:NLRextract $ cat environment.yml

• NLRextract.sh -> Script to run hmmsearch, hmmscan and pltNLR.r/vennNLR.r
• vennNLR.r -> Create venn diagram from the domain infromation and extract the combinations
• plotNLR.r -> Create barplots for the domains and the NLR proteins

1. Clone it

(base) 💻 daniel~ $ git clone https://github.com/Daniel-Ze/NLRextract.git

3. put the containing folder in your $PATH
4. chmod a+x NLRextract
5. Edit line 3 in NLRextract if you chose to install it anywhere else than your home folder:

1#!/bin/bash2########################################################### To get the script running3 NLRextracthome=~/NLRextract# edit this line4########################################################### and put the folder in PATH

5. Install the environment:

(base) 💻 daniel~ $cd NLRextract(base) 💻 daniel:NLRextract $ mamba env create -f environment.yml

(base) 💻 daniel:NLRextract $ NLRextract/Users/daniel/miniconda3/etc/profile.d/conda.sh exists.[info]No conda environment name supplied. Defaulting to: NRLextract[info]Activating conda environment NLRextract:[info] - Found Rscriptin your path.[info] - Found hmmsearchin your path.[info] - Found bedtoolsin your path.[error]No protein file supplied.NLRextract will run hmmersearch and searchfor theHMMs locatedin /Users/daniel/NLRextract/hmm/.Make sure that you supply a protein multifasta file.Usage: NLRextract -p protein.fa-c number of CPUs to use (default: 1) -p path to the protein multifasta (mandatory)-s suffixfor folder (default: random string)-e conda environment (default: NLRextract)

The script will generated several folder with the output of the different tools used:

NLRextract_TAIR10_test/├── [3.1K]  nlr.stderr# Error reports of all steps├── [204K]  clust/# Alignment and phylogenetic tree data├── [ 92K]  domain/# Fasta sequences of the extracted domains├── [1.1M]  fasta/# Fasta sequences of the extracted NLR proteins├── [1.2M]  gff/# GFF files of domains and NLR proteins├── [233M]  hmmer/# HMMER results for the NLR HMM motifs├── [ 59K]  name/# Sequence names of NLR proteins and domains├── [ 26K]  seqname/# Sequence names of proteins with a NLR domain└── [ 34K]  stats/# Summary of findings plus plots

UPDATE 24.02.21

• switched from awk to gawk=5.1.0

PREVIOUS UPDATES

• Added usage info
• Added suffix and number of CPU options
• Adjusted plotNLR.r output
• Added phylogenetic trees

This work is licensed under aCreative Commons Attribution 4.0 International License.

[1] Hiroaki Adachi, Mauricio P Contreras, Adeline Harant, Chih-hang Wu, Lida Derevnina, Toshiyuki Sakai, Cian Duggan, Eleonora Moratto, Tolga O Bozkurt, Abbas Maqbool, Joe Win, Sophien Kamoun, 2019, An N-terminal motif in NLR immune receptors is functionally conserved across distantly related plant species. eLife, 8:e49956http://dx.doi.org/10.7554/eLife.49956
[2] Burkhard Steuernagel, Florian Jupe, Kamil Witek, Jonathan D.G. Jones, Brande B.H. Wulff, 2015, NLR-parser: rapid annotation of plant NLR complements. Bioinformatics, Vol. 31, Issue 10, Pages 1665–1667https://doi.org/10.1093/bioinformatics/btv005
[3] Jiorgos Kourelis, Toshiyuki Sakai, Hiroaki Adachi, Sophien Kamoun, 2021, RefPlantNLR: a comprehensive collection of experimentally validated plant NLRs. bioRxiv,https://doi.org/10.1101/2020.07.08.193961