RRGPredictor, a set-theory-based tool for predicting pathogen-associated molecular pattern receptors (PRRs) and resistance (R) proteins from plants

In plant-pathogen interactions, plant immunity through pathogen-associated molecular pattern receptors (PAMPs) and R proteins, also called pattern recognition receptors (PRRs), occurs in different ways depending on both plant and pathogen species. The use and search for a structural pattern based on the presence and absence of characteristic domains, regardless of their disposition within a sequence, could be efficient in identifying PRRs proteins. Here, we develop a method mainly based on text mining and set theory to identify PRR and R genes that classify them into 13 categories based on the presence and absence of the main domains. Analyzing 24 plant and algae genomes, we showed that the RRGPredictor was more efficient, specific and sensitive than other tools already available, and identified PRR proteins with variations in size and in domain distribution throughout the sequence. Besides an easy identification of new plant PRRs proteins, RRGPredictor provided a low computational cost.