Gene data processing method and gene data processing device

Abstract

The embodiment of the invention discloses a gene data processing method and a gene data processing device. The method comprises the following steps: receiving gene data of a specified feature type of a reference population; preprocessing the gene data to obtain standardized gene data; performing feature gene selection on the standardized gene data by a LASSO method to obtain feature gene data; upon a cross validation method, dividing a sample set of the feature gene data into test samples and training samples; injecting the training samples into a classifier to obtain a trained classifier; injecting the test samples into the trained classifier; performing feature classification on the test samples, and performing statistics on the classification accuracy of the classifier. According to the gene data processing method and the gene data processing device provided by the embodiment of the invention, the accuracy of the feature gene selection can be improved, and the influence of the selection of the test samples and the training samples on the classification accuracy is lowered.

Description

technical field [0001] The invention relates to the technical field of gene data processing, in particular to a gene data processing method and device. Background technique [0002] DNA microarray (gene chip) technology is a major technological breakthrough in the field of molecular biology, and is widely used in various fields of biology and medical research, such as large-scale DNA sequencing, disease diagnosis, gene regulation and interaction relationship mining, etc. . Least Absolute Shrinkage and Selection Operator (LASSO) is a feature selection method based on a paradigm, which is used to describe a class of constrained optimization problems. The basic idea is to minimize the sum of squared residuals under the constraint that the sum of the absolute values ​​of the regression coefficients is less than a constant, so that some regression coefficients that are strictly equal to 0 can be generated and an interpretable model can be obtained. [0003] The expression level...

AI Technical Summary

Problems solved by technology

In the current commonly used gene data processing methods, the coverage rate of sample gene data selection is low, the accuracy of feature gene selection is low, and the selection of test samples and training samples has a greater impact on classification accuracy.

Images