Application of counter propagation artificial neural network for classification and genetic diversity assessment of somePseudomonasspecies

Abstract

Members of the genus Pseudomonas bacterium are of great interest because of their importance in plant disease. In this study, DNA fingerprints of 60 strains of Pseudomonas bacteria including three species of Pseudomonas syringae (Pseudomonas syringae pv. syringae (Pss) and Pseudomonas syringae pv. Lachrymans (Psl)), Pseudomonas savastanoi (Psa) and Pseudomonas tolaasii (Pt) were used for developing a robust predictive classification model. The DNA fingerprints were obtained by repetitive polymerase chain reaction (Rep-PCR) using enterobacterial repetitive intergenic consensus (ERIC), repetitive extragenic palindromes (REP), and BOXAIR primers. The classification results of counter propagation artificial neural network (CP-ANN) modeling indicated that a combination of Rep-PCR fingerprinting and chemometrics analysis can be used as an effective and powerful methodology to differentiate species of Pseudomonas and pathovars of P. syringae strains based on a predictive model.