A Convolutional Neural Network-Based Approach to Identify the Origins of Replication in Saccharomyces Cerevisiae *

摘要: DNA replication is key to the inheritance of genetic information. Accurate, efficient and rapid identification origins (ORIs) crucial for understanding mechanism replication. Especially eukaryotes, each their gene sequences contains multiple ORIs more Although there are many predictors designed identify eukaryotes’ ORIs, them only targeted with a fixed length. In addition, prediction accuracies not satisfying, which still has great room be improved. view limitations in this field, convolutional neural network-based approach developed study different lengths Saccharomyces cerevisiae (S. cerevisiae). As combining field Natural Language Processing (NLP), trinucleotide feature vectors constructed by Word2vec represent so as subsequent using Text-Convolutional Neural Network. result, overall success rate 88.3% was achieved proved effeciency proposed method any