What is PRGminer?

A powerful deep learning tool for identifying and classifying plant resistance genes (R-genes) in plant genomes.

PRGminer
98%
Accuracy
~2 min
Processing Time
Highly Efficient
Scalability

High Accuracy

Advanced machine learning models trained on extensive protein sequence data for precise R-gene identification.

Fast Processing

Efficient analysis of large protein sequence datasets with optimized computational pipelines.

Two-Phase Analysis

Comprehensive analysis with initial R-gene identification followed by detailed classification.

How Plant Resistance Works

  • When activated, these genes trigger a cascade of molecular processes
  • Produces antimicrobial compounds for defense
  • Strengthens cell walls to prevent pathogen invasion
  • Initiates programmed cell death in infected areas

Two-Phase Analysis

Phase I: R-gene Prediction

Predicts input protein sequences as R-genes or non-R-genes with >98% accuracy

Phase II: R-gene Classification

Classifies R-genes into eight different classes with >98% accuracy with 5-fold cross validation.

Plant Resistance Gene Classes

Coiled-coil-NBS-LRR (CNL)

Contains a central nucleotide-binding (NB) subdomain as part of a larger entity called the NB-ARC domain. C-terminal to the NB-ARC domain lies a leucine-rich repeat (LRR) domain, which is sometimes followed by an extension of variable length. Hence, this group of R proteins is collectively referred to as NB-LRR proteins. If N-terminal region contain a predicted coiled-coil structures (CC), non-TIR NB-LRR proteins are collectively referred to as CC-NB-LRR or CNL.

Kinase (KIN)

Kinase domain involved in resistance process

Lectin receptor-like kinase (LECRK)

Contains domains LECM, Kinase, and might also contain a TM domain.

Lysin motif receptor kinase (LYK)

Contains domains LYSM, Kinase, and might also contain a TM domain.

Receptor like protein (RLP)

Receptor Like Proteins consists of a leucine-rich receptor-like repeat, a transmembrane region of ~25 AA, and a short cytoplasmic region, with no kinase domain. The extracellular leucine rich repeat (eLRR) shows high homology with the RLKs eLRR. RLPs can recognise avirulence genes in a indirect way like Cf2 that recognise the avirulence gene Avr2 through the cysteine protease Rcr3. There are also some hypotheses about a direct recognition, but this has not yet been demonstrated. RLPs lack an obvious signalling domain, suggesting that defence response activation is mediated through interactions with other partner proteins. Activation of RLP typically results in a rapid accumulation of active oxygen species (AOS), changes in cellular ion fluxes, activation of protein kinase cascades, changes in gene expression and, possibly, targeted protein degradation.

Receptor-like kinase (RLK)

RLKs, or Receptor like Kinases, consist of an extracellullar leucine-rich repeat region (eLRR) that consist of 25-38aa conferring broad interaction surface that is well suited to interact with multiple ligands and an intracellular kinase domain. The eLRR domain plays the recognising role while the kinase triggers the downstream activation cascades. In Arabidopsis genome RLKs constitute a large gene family divided in 44 subclasses, 12 of them have the extracellular domain LRR while the other use different type of receptor like B-lactin and many others.

Toil-inteleukin receptor domain (TIR)

Contains TIR domain only, lack of LRR or NBS

TIR-NB-LRR (TNL)

Contains a central nucleotide-binding (NB) subdomain as part of a larger entity called the NB-ARC domain. C-terminal to the NB-ARC domain lies a leucine-rich repeat (LRR) domain, which is sometimes followed by an extension of variable length. Hence, this group of R proteins is collectively referred to as NB-LRR proteins. If N-terminal region shows homology to the protein domain Interleukin-1 Receptor (IL-1R), called the TIR domain, these proteins are referred to as TIR-NB-LRR or TNL.