Prediction Of Type ? Secreted Effectors Based On Word Embedding And Deep Learning

The type ? secreted effectors(T3SEs)are one of the indispensable proteins in the growth and reproduction of Gram-negative bacteria.In particular,the pathogenesis of Gram-negative bacteria depends on the type ? secreted effectors,and by injecting T3 SEs into a host cell,the host cell's immunity can be destroyed.The high diversity of T3SE sequences and the lack of defined secretion signals make it difficult to identify and predict.Moreover,the related study of the pathological system associated with T3SE remains a hot topic in bioinformatics.Some computational tools have been developed to meet the growing demand for the recognition of T3SEs and the studies of type III secretion systems(T3SS).Although these tools can help biological experiments in certain procedures,there is still room for improvement,even for the current best model,as the existing methods adopt hand-designed feature and traditional machine learning methods.In this study,we propose two T3SE prediction models.The first model is a predictor model based on word embedding and deep learning,called WEDeepT3.Our work consists of three key steps.First,we train protein word vectors in a large-scale protein sequence corpus.Second,we combine word embeddings with traditional protein features(PSSM)to build a more comprehensive feature representation.Finally,we build a deep neural network model for predicting T3SE.The second model is a self-attention-based prediction model.Our work involves two key steps.First,we build a large-scale protein sequence corpus and pre-train the language model on the corpus.Then,we fine-tune the parameters of the saved model on the T3SE dataset.Both prediction models are universal models in protein sequence research,and experimental results show that the performance of both models exceeds the existing T3SE prediction models.