Abstract
AbstractBackgroundThe pathogen of bananaFusarium oxysporumf. sp. cubense race 4(Foc4) infects almost all banana species, and it is the most destructive. The molecular mechanism of the interactions betweenFusarium oxysporumand banana still needs to be further investigated.MethodsWe use both the interolog and domain-domain method to predict the protein–protein interactions (PPIs) between banana and Foc4. The predicted protein interaction sequences are encoded by the conjoint triad and autocovariance method respectively to obtain continuous and discontinuous information of protein sequences. This information is used as the input data of the neural network model. The Long Short-Term Memory (LSTM) neural network five-fold cross-validation and independent test methods are used to verify the predicted protein interaction sequences. To further confirm the PPIs between banana and Foc4, the GO (Gene Ontology) and KEGG (Kyoto Encylopedia of Genes and Genomics) functional annotation and interaction network analysis are carried out.ResultsThe experimental results show that the PPIs for banana and foc4 predicted by our proposed method may interact with each other in terms of sequence structure, GO and KEGG functional annotation, and Foc4 protein plays a more active role in the process of Foc4 infecting banana.ConclusionsThis study obtained the PPIs between banana and Foc4 by using computing means for the first time, which will provide data support for molecular biology experiments.
Funder
National Natural Science Foundation of China
Natural Science Foundation of Guangxi Province
Département Caractérisation et Élaboration des Produits Issus de l’Agriculture, Institut National de la Recherche Agronomique
Publisher
Springer Science and Business Media LLC
Subject
Molecular Biology,Biochemistry
Cited by
6 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献