Method and systems for prediction of HLA class ii-specific epitopes and characterization of cd4+ t cells

A cell and prediction model technology, applied in biological systems, chemical instruments and methods, used to analyze two-dimensional or three-dimensional molecular structures, etc., can solve problems that have not yet been fully characterized

Pending Publication Date: 2021-10-01
BIONTECH US INC
View PDF69 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Endogenous processing and presentation of HLA class II ligands is a complex process involving...

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
  • Method and systems for prediction of HLA class ii-specific epitopes and characterization of cd4+ t cells
  • Method and systems for prediction of HLA class ii-specific epitopes and characterization of cd4+ t cells
  • Method and systems for prediction of HLA class ii-specific epitopes and characterization of cd4+ t cells

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0622] Example 1. HLA II class binding predictor performance

[0623] In this embodiment, the proven data set containing the observed mass spectromethope peptide and the bait peptide is used to analyze the combination predictor neonmhc2 (neon) and NETMHCIPAN. Performance Figure 4 ). For the Neon binding predictor, a separate model is constructed for each MHC II allele displayed. The height of the strip shows a positive predictive value (PPV). When the allele is predicted, the allele is sorted by the performance of the model. Compared to NetMHCIPAN, Neon binding predictors show higher PPV in all alleles.

[0624] In this embodiment, the effect of SPI threshold on the verification of the combined predictor is also tested. Figure 5 ). The HLA Group II class combines the performance of the predictor when training / verification is performed on a group peptide having different score intensity (SPI) cutoffs. Different SPI cutoff conditions: Use the observed MS hit peptide of 70SPI to tr...

Embodiment 2

[0629]Example 2. Neural Network Architecture

[0630] In this embodiment, a neural network is used to obtain a training algorithm ( Figure 9 ). The input peptide is represented as a 20 polymer, and a shorter peptide is filled with "missing" character. Each peptide has 31 dimensional embedded, so input to the neural network is a 20x31 matrix. Before being treated by the neural network, the 20x31 matrix is ​​normalized according to the feature value average and the standard deviation in the training set. The first convolution layer has 9 amino acids and 50 filters (also referred to as channels), with RELU activation functions. Next, it is normalized, then the space is discarded, and the discard rate is 20%. Next, another convolution layer has 3 cores and 20 filters, and the RELU activation function, and then the batch normalization and spatial discard, the discard rate is 20%. The global maximum pool is then applied, and the maximum activated neurons are obtained in each filter in 2...

Embodiment 3

[0631] Example 3. Scalable scheme for the analysis of mono-alleles MHC II ligand

[0632] The knowledge of the MHC II-based binding motif can be calculated based on two in vitro binding tests, and the EC50 is calculated using a cell MHC, and another use of purified MHC calculated IC50. The leading HLA II prediction algorithm NETMHCIPAN is specially trained on this data.

[0633] Limited number of human HLA II alleles currently get more than 200 examples of confirmed combined peptide (affinity Figure 12e ) They are almost a 15 polymer. These experiments cover only the most common colors HLA-DR alleles, limited to non-Caucasic group specific alleles (for example, HLA-DRB1 * 15: 02) coverage, and there is almost no cover of common HLA-DP And HLA-DQ alleles. The current HLA II prediction performance, even in the common high Caucasian allele, it is clearly lagging behind the accuracy of the MHC I class; the ROC curve is only slightly better than the random curve.

[0634] Considering t...

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 present disclosure provides method for preparing a personalized cancer vaccine. The present disclosure also provides a method to train a machine-learning HLA-peptide presentation prediction model.

Description

[0001] cross reference [0002] The present application claims Temporary application for US Temporary Application No. 62 / 783, 914, filed on December 21, 2018, US Temporary application submitted on March 29, 2019, filed on May 31, 2019, No. 62 / 855, 379, 2019 The US temporary application and the US temporary application of US temporary applications, and each of the temporary applications is incorporated herein by reference. Background technique [0003] The main tissue compatibility complex (MHC) is a gene complex encoding a human leukocyte antigen (HLA) gene. The HLA gene was expressed as a protein isopolymer having a circulating T cell on a human cell surface. HLA genes are highly polymorphic, allowing their fine-tuning adaptive immune systems. Adaptive immune response partially depends on T cell identification and eliminates the ability to exhibit disease-associated peptide antigen from human leukocyte antigen (HLA) isotactic cells. [0004] In humans, endogenous and exogenous pr...

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): G16B20/00
CPCG16B40/20G16B15/30G16B40/10G16B20/30G16B25/10A61P35/00A61K39/39C07K16/2833G16B30/00G16B40/00G16B5/00G16B30/10C07K14/70539C12N15/63
Inventor 迈克尔·史蒂文·鲁尼詹尼弗·格雷斯·阿贝林杜米尼克·巴特尔梅罗伯特·卡门
Owner BIONTECH US INC
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