Genetic and environmental correlation-based colorectal cancer data model analysis method

Abstract

The invention provides a genetic and environmental correlation-based colorectal cancer data model analysis method. The method comprises the steps of receiving colorectal cancer (CRC) data of a specified feature type of a reference population; preprocessing the data to obtain standardized data; based on the standardized data, classifying the data; performing feature selection on each subclass by using a sparse principal component analysis method, an information entropy method and/or a Relief method; obtaining an intersection of three methods by using a Venn diagram, and obtaining features with remarkable difference by using a U test; and dividing a feature gene data sample set into test samples and training samples, obtaining a trained classifier according to the training samples, injecting the test samples into the trained classifier, performing feature classification on the test samples, and performing statistics on classification accuracy of the classifier. According to the method, the accuracy of extracting carcinogens can be improved and the classification accuracy can be improved.

Description

technical field [0001] The present invention relates to the technical field of bioinformatics, mainly relates to methods of biological data analysis and biological data mining, and specifically relates to the establishment of a robust colorectal cancer data model based on large genetic and environment-related colorectal cancer data, and based on the data model data analysis and mining. Background technique [0002] Colorectal cancer, including colon and rectal cancer, is a leading cause of cancer-related morbidity and mortality worldwide. In 2002, there were approximately 1,023,152 newly diagnosed cases of colorectal cancer, and 528,978 patients died of colorectal cancer. Colorectal cancer ranks fourth in the incidence and death spectrum of malignant tumors in men, and ranks fourth in malignant tumors in women. It ranks third in the incidence spectrum and fifth in the death spectrum. That is to say, one person is newly diagnosed with colorectal cancer every half minute, an...

AI Technical Summary

Problems solved by technology

However, the above research methods have certain limitations and have not improved the accuracy of predicting colorectal cancer

Images