Method for semi-supervised learning prediction of protein secondary structure

Abstract

The invention discloses a method for semi-supervised learning prediction of a protein secondary structure. The method comprises the following steps: (1) acquiring a protein sequence data set; (2) performing data cleaning and feature extraction on the acquired data set; (3) establishing a Simi-GAN neural network model; (4) training the Simi-GAN neural network model; (5) adjusting parameters of theSimi-GAN neural network model; (6) evaluating the Simi-GAN neural network model. According to the method, a semi-supervised prediction model can be established for the protein secondary structure under the condition that a large number of missing value labels exist, thereby saving a large amount of manpower and financial resources.

Description

technical field [0001] The present invention relates to the fields of bioinformatics and deep learning, and uses a deep learning model to perform semi-supervised learning and prediction of protein secondary structure, which belongs to a key research issue of bioinformatics prediction. Specifically, it relates to a method for semi-supervised learning and prediction of protein secondary structure, which uses protein data sets with missing labels to train a deep learning classification model to predict protein secondary structure. Background technique [0002] Protein secondary structure prediction is the inference of the secondary structure of protein fragments based on their amino acid sequences. In bioinformatics and theoretical chemistry, protein secondary structure prediction is important for medicine and biotechnology, such as drug design and the design of novel enzymes. Since secondary structures can be used to find long-range relationships of proteins with non-alignabl...

AI Technical Summary

Problems solved by technology

However, labeling protein secondary structure labels requires a lot of manpower and financial resources, and even time.

Images