Application Of LSTM In Protein Structure Prediction LINA

Abstract

In this paper the authors discuss the applications of LSTM Neural Network in Protein Structure Prediction. The main idea is to construct a LSTM neural network. Predicting the secondary structure of a protein is the basis content for predicting its spatial structure. In this article, a position-specific scoring matrices (PSSM) containing evolutionary information is linked to other features to construct a completely new feature set. The CB513 data set is selected to construct LSTM neural networks to predict the secondary structure of the sequence. Experiments have shown that the proposed method effectively improves the prediction accuracy and is better than the previous method. The idea in this paper can also be applied to the analysis of other sequences.