Information recognition of pathogenic modules in gene statistics of big data

Abstract

Aiming at the problem that the selection of the F value is too small when the negative outliers are removed by traditional recognition methods, a recognition method of pathogenicity module information in gene statistics of high-dimensional big data is proposed. This method involves using gene chips to obtain gene expression data, constructing a dynamic network to screen pathogenic module genes, preprocessing gene expression data, calculating the maximum information coefficient characteristics of pathogenic module information by using a feature matrix, standardizing the processing of pathogenic module information data, establishing pathogenic module information recognition rules and completing pathogenic module information in gene statistical interest recognition of high-dimensional big data interest. The experimental results show that compared with the traditional recognition methods, the disease module information recognition method in high-dimensional big data gene statistics is less affected by the K value and the actual recognition accuracy is up to 98%.