miRNA-disease association relation prediction method based on similarity and logical matrix decomposition

Abstract

The invention discloses a miRNA-disease association relation prediction method based on similarity and logical matrix decomposition. The method comprises the steps that firstly, the disease function similarity and miRNA function similarity are calculated; then, by utilizing the known miRNA-disease association relation, the disease gaussian kernel similarity and miRNA gaussian kernel similarity areestablished; the disease function similarity and gaussian kernel similarity are integrated to obtain the final disease similarity, and the miRNA function similarity and gaussian kernel similarity areintegrated to obtain the final miRNA similarity. Finally, based on a logical matrix decomposition model, potential characteristic vectors of miRNA and diseases are predicted, and the association relation fraction of the miRNA and disease pairs is calculated according to the logical regression function. According to the miRNA-disease association relation prediction method, the relation of the newmiRNA diseases can also be predicted, the consumed labor, material resources and financial resources in a biochemistry laboratory are avoided, and the accuracy is high.

Description

technical field [0001] The invention belongs to the field of systems biology and relates to a miRNA-disease correlation prediction method based on similarity and logical matrix decomposition. Background technique [0002] More and more studies have shown that miRNA, as a kind of non-coding RNA with a length equal to about 22nt, plays a very important role in many complex human diseases. Therefore, identifying the association between miRNA and disease will greatly promote the understanding of disease mechanism, drug development, and disease treatment. While the identification of miRNA-disease interactions by bioassays is notoriously expensive, time-consuming and challenging, computational modeling can provide a low-cost, high-efficiency approach to miRNA-disease relationships Make predictions. Therefore, with the development of computing technology, relatively many algorithms have emerged to predict the relationship between miRNA and disease. [0003] At present, there are...

AI Technical Summary

Problems solved by technology

[0011] Although the above methods have achieved some good results in predicting the miRNA-disease relationship and provided an important basis for medical research, there are still some shortcomings. Experiments, network-based forecasting methods and other machine learning-based forecasting methods still need further improvement in forecasting accuracy

Images