Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Tumor neoantigen prediction method and neoantigen prediction system based on deep learning model

An antigen, a new technology, applied in the field of biomedicine, can solve the problems of low accuracy and poor efficacy of anti-tumor vaccines, and achieve the effects of high precision, improved effectiveness and reduced cost

Pending Publication Date: 2022-04-12
格源致善(上海)生物科技有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] (2) Low accuracy rate: the accuracy rate of this prediction method is about 30%-40%
[0010] Therefore, the accuracy of the reported tumor neoantigen prediction methods is low, which in turn leads to poor efficacy of neoantigen-based anti-tumor vaccines

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
  • Tumor neoantigen prediction method and neoantigen prediction system based on deep learning model
  • Tumor neoantigen prediction method and neoantigen prediction system based on deep learning model
  • Tumor neoantigen prediction method and neoantigen prediction system based on deep learning model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0118] The present invention provides a neoantigen prediction system, which includes a sample collection device, a transcriptome sequencing data analysis module, and a whole exome sequencing data analysis module (including somatic cell non-synonymous mutation analysis, mutant peptide chain and flanking sequence acquisition, HLA typing analysis unit), neoantigen prediction device.

[0119] The sample collection device includes a tumor cell and a normal cell collection device, and transports the sample to the transcriptome sequencing data analysis module and / or the whole exome sequencing data analysis module, and the tumor cells perform transcriptome sequencing and whole exome sequencing of the sample respectively , normal cells were subjected to whole exome sequencing. The whole exome sequencing data analysis module receives and compares the whole exome sequencing data of tumor cells and normal cells, calculates and presents the somatic cell non-synonymous mutation data to the ...

Embodiment 2

[0127] The construction of the neural network of the neoantigen prediction system is mainly divided into two steps:

[0128] (1) Training data acquisition

[0129] Our training data acquisition methods are as follows: figure 1 shown. We constructed cell lines with high-frequency HLA subtypes from the Chinese population. Firstly, specific primers for HLA-A, HLA-B and HLA-C were designed, and PCR was used to amplify B-LCL cells ( CRL-2369 TM ) in HLA-A, HLA-B and HLA-C gene fragments, and then these gene fragments were subcloned into retroviral vectors, and finally LCL 721.221 cell line (human HLA class I deletion cell line)( CRL-1855 TM) to obtain HLA subtype cell lines. Use protein immunoprecipitation and mass spectrometry to obtain peptide chain sequences that can bind to a specific HLA molecule, select peptide chains with a length of 8-11 amino acids, and fill in 11 peptide chains with a length of amino acids less than 11, and intercept The 5 amino acids on the lef...

Embodiment 3

[0133] The process of neoantigen prediction is as follows: image 3 shown. First, collect the patient's tumor tissue and peripheral blood samples, extract DNA, and perform whole-exome sequencing. Firstly, use FastQ software to perform QC (quality control) processing on the sequencing data, and then use BWA software to compare and splicing the sequencing data with the reference genome. , using Mutect2 software to analyze Somatic mutations (somatic mutations) in tumor-blood paired samples, including SNV (single-nucleotide variant, single nucleotide mutation), InDel (insertion / deletion mutation, insertion / deletion mutation), frameshift ( frameshift mutation), etc., screen for non-synonymous mutations, and generate mutant peptide chains and their flanking sequences. The exome sequencing data of peripheral blood samples were analyzed by xHLA software for HLA typing. RNA was extracted from tumor tissue, and transcriptome sequencing was performed to obtain gene expression level (TP...

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

No PUM Login to View More

Abstract

The invention belongs to the field of biological medicine, and discloses a neoantigen prediction method, which comprises the following steps: collecting a sample of a neoantigen to be predicted, extracting genome DNA and RNA of the sample, carrying out whole exon sequencing and transcriptome sequencing, carrying out HLA typing analysis according to whole exon sequencing data of the sample, and carrying out RNA expression level detection according to transcriptome sequencing data. Comparing and splicing whole exon sequencing data with a human reference genome, and analyzing somatic mutation of tumor-normal paired samples to obtain a mutant peptide chain sequence and a flanking sequence thereof; and inputting the HLA typing, the mutant peptide chain sequence and the flanking sequence thereof, and the gene expression level value into a deep learning model to obtain a predicted neoantigen. The invention further provides a neoantigen prediction system, a corresponding device and application. The accuracy of neoantigen prediction can be remarkably improved.

Description

technical field [0001] The invention belongs to the field of biomedicine, and relates to a method for predicting tumor neoantigens, in particular to a method for predicting tumor neoantigens based on a deep learning network, a prediction system, a device and applications thereof. Background technique [0002] Today, tumor immunity has become one of the hottest tracks. However, from the perspective of clinical efficacy, the road to tumor immunity is long and difficult. Taking PD-1 / PD-L1 as an example, only 20%-30% of tumor patients can benefit from it; while CAR-T is only effective for blood tumors, especially B lymphocyte tumors, and has serious side effects. There are no effective treatment options for most cancer patients, and more possibilities need to be explored for tumor immunotherapy, and personalized tumor vaccines are one of them. The development of personalized tumor vaccines is an integrated technology that combines precise genetic testing and tumor immunotherap...

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
IPC IPC(8): G16B25/10G16B30/10G06N3/04G06N3/08
Inventor 李锐雷俊卿虞韩川枝秦汉楠苏小平李伟迎
Owner 格源致善(上海)生物科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products