Resistify 🍃

More than 2,500 downloads - thank you all!

Resistify is a program which rapidly identifies and classifies plant resistance genes from protein sequences. It is designed to be lightweight and easy to use.

What's new in v0.6.0?

The release of v0.6.0 has brought a number of changes to Resistify. First, you'll note that there are now two modes available - NLR and PRR - which identify NLRs and PRRs respectively.

The NLR pipeline is largely the same, but has received multiple performance improvements which should allow it to utilise more threads simultaneously and significantly reduce memory usage. As a result of these changes, the --threads mode has now been removed which was a bit of a lie anyway, as numpy would use them all regardless. The --ultra setting has been renamed as --retain.

The PRR pipeline is new to Resistify and is currently in development. It uses a re-implementation of TMbed to predict transmembrane domains, from which it will identify and classify RLP/RLKs according to a recently described classification system. Feel free to give it a try and offer suggestions! Due to other commitments I can't currently benchmark this properly and make no guarantees to its accuracy yet.

Installation

Resistify is available via the Bioconda channel:

conda create -n resistify bioconda::resistify
conda activate resistify

When using conda, please ensure that your Bioconda has been configured correctly.

Docker/Podman containers are also available through the biocontainers repository. To use these with - for example - singularity, simply run:

singularity exec docker://quay.io/biocontainers/resistify:<tag-goes-here>

If you are having issues with conda, you can instead try installing directly from the repository:

git clone https://github.com/SwiftSeal/resistify.git
cd resistify
pip install .

Note that resistify requires hmmer to be installed and available in your system's PATH, which will not be installed automatically when using pip.

Usage

Identifying NLRs

To predict NLRs within a set of protein sequences, simply run:

resistify nlr <input.fa>

and Resistify will identify and classify NLRs, and return some files:

• results.tsv - A table containing the primary results of Resistify.
• motifs.tsv - A table of all the NLR motifs identified for each sequence.
• domains.tsv - A table of all the domains identified for each sequence.
• annotations.tsv - A table of the raw annotations for each sequence.
• nbarc.fasta - A fasta file of all the NB-ARC domains identified.
• nlr.fasta - A fasta file of all NLRs identified.

By default, Resistify will only return sequences with NB-ARC domains. If you wish to identify highly fragmented NLRs, you can use the --retain option which will predict and report NLR-associated motifs for all sequences. It'll be a bit slower!

If you want to increase the sensitivity of coiled-coil domain annotation, you can use the option --coconat. This will use CoCoNat to predict coiled-coil domains. In practice, I wouldn't expect this mode to pick up on a significant number of missed CC domains, but it can pick up on cryptic CCs that do not have an identifiable EDVID motif.

How does it work?

Resistify carries out an initial search for common NLR domains to quickly filter and annotate the input sequences. Then, Resistify executes a re-implementation of NLRexpress to conduct a highly accurate search for NLR-associated motifs. If --coconat is used, this will also be executed to scavenge for potentially missed coiled-coil domains.

Identifying PRRs

To predict PRRs within a set of protein sequences, simply run:

resistify prr <input.fa>

and Resistify will identify and classify PRRs, and return some files:

• results.tsv - A table containing the primary results of Resistify.
• motifs.tsv - A table of all the LRR motifs identified for each sequence.
• domains.tsv - A table of all the domains identified for each sequence.
• annotations.tsv - A table of the raw annotations for each sequence.

• prr.fasta - A fasta file of all PRRs identified.

  This mode uses TMBed to predict transmembrane domains. It greatly benefits from GPU acceleration - running with CPUs only will be extremely slow and memory intensive.

Downloading model data

By default, Resistify will automatically download the models required for CoConat and TMbed to your $HOME/.cache directory. If you'd like to manually install the databases instead, you can use the resistify download_models utility to download these to a directory of your choice. To provide these local models to the CoCoNat and TMbed processes, simply pass the path of the models directory via the --models argument. Approximately 13G of disk space is required. If you only intend to use the NLR module without --coconat, no external databases will be downloaded.

Results

results.tsv (nlr)

The main column of interest is "Classification", where we can see that it has been identified as a canonical CNL. The "Motifs" column indicates the series of NLR-associated motifs identified across the sequence - this can be useful if an NLR has an undetermined or unexpected classification. The columns "MADA", "MADAL", and "CJID" correspond to common NLR sequence signatures. Here, it appears that ZAR1 has a MADA-like motif.

results.tsv (prr)

For PRRs, sequences can be of the type RLP or RLK - both are single pass transmembrane proteins, and RLKs have an internal kinase domain. Classification refers to the domains identified in the external region. If multiple domains are identified, they will each be reported as a semi-colon separated list. If a signal peptide is identified in the sequence, this is reported accordingly.

motifs.tsv

Here, the positions, probabilities, and sequence of NLRexpress motif hits are listed. The five amino acids upstream and downstream of the motif site are also provided. In PRR mode, only LRR motifs will be reported.

domains.tsv

This file contains the coordinates of the domains identified by Resistify.

annotations.tsv

This file contains the raw annotations for each sequence, and the method which was used to identify them.

Output visualisation

I've kept the output files of Resistify fairly minimal so that users can carry out their own analysis/visualisation. Here are some examples of how Resistify can be used to create basic plots.

Phylogenetics

Resistify extracts the NB-ARC domains of each hit so we can easily build a phylogenetic tree. Here, we create a tree rooted on the NB-ARC domain of CED-4. The mafft | fastree method is used here for brevity rather than accuracy.

echo -e ">ced4\nREYHVDRVIKKLDEMCDLDSFFLFLHGRAGSGKSVIASQALSKSDQLIGINYDSIVWLKDSGTAPKSTFDLFTDILLMLARVVSDTDDSHSITDFINRVLSRSEDDLLNFPSVEHVTSVVLKRMICNALIDRPNTLFVFDDVVQEETIRWAQELRLRCLVTTRDVEISNAASQTCEFIEVTSLEIDECYDFLEAYGMPMPVGEKEEDVLNKTIELSSGNPATLMMFFKSCEPKTFEKMAQLNNKLESRGLVGVECITPYSYKSLAMALQRCVEVLSDEDRSALAFAVVMPPGVDIPVKLWSCVIPVD" >> output/nbarc.fasta

mafft output/nbarc.fasta | fasttree > output/nbarc.tree

We can now plot the tree:

library(tidyverse)
library(ggtree)

tree <- read.tree("output/nbarc.tree")
tree <- treeio::root(tree, outgroup = "ced4")

results <- read_tsv("output/results.tsv") |>
  mutate(Sequence = paste0(Sequence, "_1"))

myplot <- ggtree(tree, layout = "circular") %<+% results

myplot <- myplot +
  geom_tippoint(aes(colour = Classification))

Domain plotting

Sometimes, it might be of interest to plot the distribution of domains and motifs across each NLR. Achieving this with Resistify is quite simple:

library(tidyverse)

motif_translation = c(
  "extEDVID" = "CC",
  "bA" = "TIR",
  "aA" = "TIR",
  "bC" = "TIR",
  "aC" = "TIR",
  "bDaD1" = "TIR",
  "aD3" = "TIR",
  "VG" = "NB-ARC",
  "P-loop" = "NB-ARC",
  "RNSB-A" = "NB-ARC",
  "Walker-B" = "NB-ARC",
  "RNSB-B" = "NB-ARC",
  "RNSB-C" = "NB-ARC",
  "RNSB-D" = "NB-ARC",
  "GLPL" = "NB-ARC",
  "MHD" = "NB-ARC",
  "LxxLxL" = "LRR"
)

domains <- read_tsv("output/domains.tsv")
results <- read_tsv("output/results.tsv")
motifs <- read_tsv("output/motifs.tsv") |>
  mutate(Domain = motif_translation[Motif])

myplot <- ggplot() +
  geom_segment(data = results, aes(y = Sequence, yend = Sequence, x = 0, xend = Length)) +
  geom_segment(data = domains, aes(y = Sequence, yend = Sequence, x = Start, xend = End, colour = Domain)) +
  geom_point(data = motifs, aes(y = Sequence, x = Position, colour = Domain))

Cute! NB: Some false-positive motif hits are evident in this example - it might be of interest to not plot them, or plot only LRR motifs which tend to be a bit more informative.

Contributing

Contributions are greatly appreciated! If you experience any issues running Resistify, please get in touch via the Issues page. If you have any suggestions for additional features, get in touch!

Citing

Resistify - A rapid and accurate annotation tool to identify NLRs and study their genomic organisation
Moray Smith, John T. Jones, Ingo Hein
bioRxiv 2024.02.14.580321; doi: https://doi.org/10.1101/2024.02.14.580321

You must also cite:

NLRexpress—A bundle of machine learning motif predictors—Reveals motif stability underlying plant Nod-like receptors diversity
Martin Eliza C., Spiridon Laurentiu, Goverse Aska, Petrescu Andrei-José
Frontiers in Plant Science 2022; doi: https://doi.org/10.3389/fpls.2022.975888

If you use the CoCoNat module, please cite:

CoCoNat: a novel method based on deep learning for coiled-coil prediction
Giovanni Madeo, Castrense Savojardo, Matteo Manfredi, Pier Luigi Martelli, Rita Casadio
Bioinformatics 2023; doi: https://doi.org/10.1093/bioinformatics/btad495

If you use the PRR module, please cite:

TMbed: transmembrane proteins predicted through language model embeddings.
Bernhofer, M., Rost, B.
BMC Bioinformatics 2022; doi: https://doi.org/10.1186/s12859-022-04873-x