MicroRNA-disease association prediction method based on multi-mode stacking automatic coding machine

Abstract

The invention discloses a microRNA-disease association prediction method based on a multi-mode stacking automatic coding machine. The method comprises: forming microRNA sequence features and disease semantic similarity features; constructing a microRNA-protein-disease network, a microRNA-mRNA-disease network and a microRNA-lncRNA-disease network, and respectively obtaining network adjacent characteristics between microRNA and protein, between disease and protein, between mRNA and lncRNA and between disease and lncRNA by using a LINE network embedding method; mining, by using a multi-mode stacking automatic coding machine, the advanced abstract features of four features (self attribute features, protein network adjacent features, mRNA network adjacent features and lncRNA network adjacent features) of microRNA and diseases, thereby reducing the time complexity of a model and improving the prediction accuracy of the model; and training and predicting the processed features by using a CatBoost classifier, and taking an average value of prediction scores of the four features as a final prediction score. According to the method, the problems of high time consumption and high cost of a traditional biological experiment method are solved, so that a better classification effect is achieved, and the potential incidence relation between microRNA and diseases is predicted with higher accuracy.

Description

technical field [0001] The invention relates to the fields of machine learning and bioinformatics, in particular to a microRNA-disease association prediction method based on a multimodal stacked autoencoder. Background technique [0002] MicroRNA (miRNA) is a small molecule non-coding RNA (-22nt), which plays an important role in cells. It is estimated that 1-4% of the genes in the human genome are miRNAs, a single miRNAs regulates up to 200 mRNAs, and miRNAs usually bind to the 3' untranslated regions (UTRs) of target mRNAs through sequence-specific base pairs, repressing target mRNAs expression, thus participating in a series of important processes in the life process. Identifying potential microRNA (miRNA) and human disease associations has been a key goal in many bioinformatics research projects, which will facilitate the treatment and prevention of human diseases, molecular tool design and personalized diagnosis. [0003] Traditional biological experiments are expensi...

AI Technical Summary

Problems solved by technology

However, the human body is a unified whole, a variety of biomolecules in human cells are coordinated to maintain life activities, and the interactions between various biomolecules are interconnected. Most of the current calculation methods only consider a single type of known miRNA-disease association information without paying more attention to associations between miRNAs and disease and other biomolecules

Images