Cancer subtyping method and device based on RNA targeted sequencing and machine learning

An RNA targeting and machine learning technology, applied in the fields of biochemical equipment and methods, genomics, instruments, etc., can solve the problem of time-consuming and laborious, and achieve the effect of high efficiency, improved accuracy and high throughput

Inactive Publication Date: 2019-11-01
元码基因科技(无锡)有限公司
View PDF9 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, magnetic resonance imaging (MRI), contrast-enhanced ultrasound and computerized tomography (CT) are commonly used, but all of them need to be interpreted based on the rich experience of clinicians, and it is relatively time-consuming and laborious, and the results are somewhat subjective.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Cancer subtyping method and device based on RNA targeted sequencing and machine learning
  • Cancer subtyping method and device based on RNA targeted sequencing and machine learning
  • Cancer subtyping method and device based on RNA targeted sequencing and machine learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0100] 1. Sample information

[0101] The RNAseq expression profile data of 914 patients with lung cancer from the TCGA database were selected as samples.

[0102] 2. Experimental steps

[0103] 1. Pretreatment:

[0104] 1.1 Establish a database by using the expression data of lung cancer in the TCGA project, use the expression data of 914 genes as features, and use the classification of cancer subtypes as labels to establish a typing database.

[0105] 1.2 Use the random forest algorithm, set the maximum number of features used by each decision tree stump to 143, use 2000 decision tree stumps, classify and train the above subtype database, and select the 100 genes with the highest weight (see Table 1) And save the model, this model is the typing prediction model of the present invention, the model takes 0.5 as the probability threshold value, promptly predicts that the sample belongs to this subtype when the probability that the sample is the subtype exceeds 0.5, and the ge...

Embodiment 2

[0136] The other steps of Example 2 are the same as those of Example 1 unless otherwise specified.

[0137] 1. Sample information

[0138] The RNAseq expression profiling data of 750 patients with renal cell carcinoma from the TCGA database were selected as samples.

[0139] 2. Experimental steps

[0140] 1. Pretreatment:

[0141] 1.1 By using the expression data of renal cell carcinoma in the TCGA project to establish a database, the expression data of 750 genes are used as features, and the classification of cancer subtypes is used as labels to establish a typing database.

[0142] 1.2 Use the random forest algorithm, set the maximum number of features used by each decision tree stump to 143, use 2000 decision tree stumps, classify and train the above subtyping database, and select the 100 genes with the highest weight (see Table 3) And save the model, this model is the typing prediction model of the present invention, the model takes 0.5 as the probability threshold valu...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

PropertyMeasurementUnit
Sensitivityaaaaaaaaaa
Login to view more

Abstract

The invention discloses a cancer subtyping method and a device based on RNA targeted sequencing and machine learning. According to the method, through RNA targeted sequencing technology, a target genearea is concentrated efficiently. Through steps of inverse transcription, database establishing and sequencing, second-generation sequencing data of the target area are obtained. Furthermore a randomforest algorithm is used for training on a TCGA dataset for obtaining a tumor typing predicting model, thereby accurately performing cancer subtyping. The method according to the invention can obtaina model for realizing lung cancer and renal cell carcinoma typing with high accuracy. The method according to the invention can reduce typing cost. Furthermore the typing speed, precision and the typing result accuracy are better than that of a traditional method.

Description

technical field [0001] The invention relates to the field of cancer subtype typing, in particular to a method for cancer subtype typing based on RNA targeted sequencing and machine learning. Background technique [0002] In order to formulate reasonable and effective individualized treatment plans for cancer patients, it is very important to accurately classify cancer subtypes and identify relevant key pathogenic genes. Pathologists need more experience to be able to classify subtypes based on symptoms, slide images, etc., but there is still a certain degree of subjectivity. Therefore, being able to classify the subtypes in an automatic way can not only save the time of pathologists, but also classify the subtypes of patients in a relatively objective form, which can ultimately have an important impact on the later treatment of patients. [0003] Studies have found that the gene expression profile of metastatic tumors is different from that of metastatic tissue, but more si...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G16B20/00C12Q1/6869
CPCC12Q1/6869G16B20/00C12Q2535/122C12Q2537/165
Inventor 杨家亮王博郎继东梁乐彬张燕香孙雪王伟伟王兴枝时淑舫田埂
Owner 元码基因科技(无锡)有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap