Protein-Protein Interactions Classification. Based on Recurrent Neural Network

      Аннотация: Proteomics is the large-scale study of proteins, particularly their structures and functions. Proteins are vital parts of living organisms, as they are the main components of the physiological metabolic pathways of cells. Most proteins function in collaboration with other proteins and one goal of proteomics is to identify which proteins interact. This is especially useful in determining potential partners in cell signaling cascades. A number of techniques have been developed for the identification and classification of protein-protein interactions. The techniques developed in past years are still far from perfect. The Jordan neural network classification model tries to overcome this problem. The Jordan Neural Network takes amino acid composition of protein pairs to classify them interacting and non-interacting. Jordan neural network classification model outperforms the other methods for protein-protein interaction classification. Jordan neural network classification model proves to be better model with higher accuracy along with improved specificity and sensitivity than the various existing techniques.

Ключевые слова: protein-protein interaction, SVM, amino acid composition, Jordan Recurrent Neural Network, SVM-KNN, Back-Propagation Neural Networks