Potato Protein Secondary Structure Prediction Based On Neural Network Research

Bioinformatics is a new subject, which adopts computers to deal with and research into bioinformation. And prediction of protein secondary structure, with a long historic task, has long been a challenge at present in the research of Bioinformatics. Any new breakthrough in this research will be helpful to knowing better the spatial structures and the function of protein. What is more, it will be an important assistant to relevant industries such as biomedical engineering, agbio-tech, etc. Aimed at this need, we develop our work on the study of protein secondary structure prediction using methods of bioinformatics. This paper's main research content and achievement as follows:(1) chooser of evaluating data about the protein secondary structure predictionThis paper firstly withdrew amino acid sequences and secondary structure from the PDB protein structure database. With a view to the complex data form in them, this paper download them by hand and save as evaluating data by the format of txt. This data has built the good foundation for protein secondary structure prediction.(2) 3 kind of amino acids encoding method comparison research (data pretreatment) In view of the orthogonal code, 5 code and the Profile code and so on 3 commonly usedcodes methods. Using the BP nerve network, this paper established the evaluation model of protein secondary structure prediction. At the same time, using this model, this paper has analyzed the influence of these 3 encoding method to precision of protein secondary structure prediction.The result indicated that, the profile encoding method that is rich in "the biological evolution information" might obtain the higher predictable result. (3) Improved BP arithemetic researchConsidering the disadvantages of standard BP arithmetic, the paper introduces several improved BP neural networks to predict protein secondary structure, and analyses the precisions of prediction. The result has proved that a high precision can be Achieved by theimproved BP network which combines genetic algorithm with method of momentum and strategy that is adopted for adjusting learning efficiency.(4) The model of potato Protein Secondary Structure based on the multi-modal neural networkThis paper proposed a multi-modal neural network predictable model that synthesizes by 7 BP neural network group in protein secondary structure prediction. Union merit of the Profile code and improved BP arithmetic, and make a forecast result ultimately with the average exporting summation 7 BP neural networks. The result has proved that the average precision which predict potato protein secondary structure from prophase 68. 2 667% enhances to70. 5778%.