Prediction Of Ion Channel Proteins And Artemisia Annua Anti-tumor Related Proteins By Bioinformatics

Ion channels are ion-permeable protein pores in all cell lipid membranes.Different ion channels have unique functions in different biological processes.A large part of these ion channels are related to the occurrence and treatment of diseases,and ion channels are known to be targets for more than 700 drugs.With the rapid development of high-throughput mass spectrometry,proteomic data are accumulating rapidly.Ion channels play an important role in cell conduction.They are closely related to many diseases and play an important role in the research of cancer and cancer.They are hot areas of clinical and scientific research.In addition,studies have shown that traditional Chinese medicine has a good effect on the treatment of"channel disease",so the rapid and accurate prediction and classification of ion channels need further study.The key point of this paper is to find eight key target proteins of Artemisia annua and tumor by using text mining technology in bioinformatics.Further research shows that these key target proteins can produce anti-tumor effect by acting on the concentration of ions in cells,so the ion channels are further predicted and classified.Then,based on the physical and chemical properties and other properties and characteristics of proteins,a classifier of ion channel proteins is constructed by using stochastic forest model and support vector machine algorithm.The classifier is used to determine whether the protein sequence is an ion channel protein,and classify the ion channel proteins into family and subfamily.Once the protein sequence is identified as an ion channel,it can be further classified.In this paper,three evaluation indicators are used:sensitivity(Sn),overall accuracy(OA)and average accuracy(AA).The model is evaluated by ten-fold cross-validation.SVMProt,k-skip-n-gram and iFeature were used to extract feature vectors from the processed ion channel data set.MRMD was used to reduce the dimension of the feature vectors.Random forests and support vector machines were used to predict and classify ion channels.The experimental results showed that the feature vectors extracted by SVMProt and k-skip-n-gram were more meaningful,and MRMD performs well in feature selection,which effectively improved the accuracy of ion channel prediction and classification.Random Forest and Support Vector Machine(SVM)have the best performance in ion channel prediction and classification.In summary,the machine learning method can effectively improve the prediction and classification effect of ion channels,and can quickly and accurately predict and classify ion channels.Especially compared with other classifiers,Random Forest and Support Vector Machine(SVM)perform very well in the prediction and classification of ion channels.The most suitable set of feature vectors was found through experiments,and the method was used to predict and classify ion channels.