Screening method and application of natural killer cell markers

By combining SH2D1B gene and its protein products with peripheral blood transcriptome data and machine learning algorithms, a natural killer cell identification model was constructed, which solved the problem of insufficient screening of natural killer cell markers in the existing technology, and achieved efficient and accurate identification and enrichment of natural killer cells.

CN119274656APending Publication Date: 2025-01-07BEIJING HUADA BIO & INFORMATION FUSION TECHNOLOGY RESEARCH CO LTD
1 Cites 0 Cited by

Patent Information

Application Number
CN202310836402.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-07-07
Publication Date
2025-01-07

AI Technical Summary

Technical Problem

The screening method for natural killer cell markers in the prior art is complex and not accurate enough, making it difficult to efficiently identify and enrich natural killer cells, and there are problems of high cost and insufficient accuracy.

Method used

The SH2D1B gene and its protein products are used as markers, combined with peripheral blood transcriptome data and machine learning algorithms, and a natural killer cell identification model was constructed through gene expression profiling analysis and logistic regression model to screen out new highly specific molecular markers.

Benefits of technology

The efficient identification and enrichment of natural killer cells was achieved. The predictive model AUC value of the new marker SH2D1B was 0.92, which was close to or even better than the existing database molecular markers, reducing costs and consumption of experimental resources, and improving accuracy and specificity. sex.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a natural killer cell marker screening method and application thereof. The screening method comprises the following steps: S1, calculating a gene expression profile of end-to-end differentiated blood cells in peripheral blood under normal physiological conditions, obtaining difference characteristics of the gene expression profiles of natural killer cells and other blood cells through comparison, and screening out candidate molecular markers; and S2, constructing a natural killer cell identification gene model by combining the candidate molecular markers with a machine learning algorithm, and screening out natural killer cell markers. According to the technical scheme, based on a large number of public peripheral blood transcriptome data sets, a machine learning method is used for screening and developing new natural killer cell molecular markers. According to the new molecular marker SH2D1B of the natural killer cells obtained through screening, the prediction model AUC value is 0.92, and the identification effect is close to or even superior to that of an existing database molecular marker.
Need to check novelty before this filing date? Find Prior Art