HLA Allele-Specific Peptide Prediction via Single-Cell Profiling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for predicting peptide binding to human leukocyte antigen (HLA) molecules are limited by their reliance on biochemical affinity measurements, leading to low throughput, biased sampling, and neglect of endogenous processing and transport processes, which complicates the identification of HLA-allele specific binding peptides.
Innovation Solution
A method involving the generation of an HLA allele-specific binding peptide sequence database through isolating and sequencing peptides from cells expressing a single HLA allele, followed by training a machine learning model with these sequences to predict HLA-allele specific binding peptides, incorporating variables such as peptide sequence, amino acid properties, and intracellular processing information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If biochemical affinity measurements are used to train prediction algorithms, then prediction accuracy for HLA binding is improved, but throughput remains low and sampling is biased
Solution Approach 1:
The patent replaces traditional biochemical affinity measurement methods with mass spectrometry-based detection. This substitution enables high-throughput, unbiased sampling of peptide-HLA complexes while maintaining accurate identification of binding peptides, thereby resolving the contradiction between measurement precision and productivity
Solution Approach 2:
The patent creates cell lines expressing single HLA alleles as simplified models of complex multi-allelic systems. This copying approach allows systematic study of allele-specific peptide binding with high throughput while preserving the essential binding characteristics, improving both productivity and sampling comprehensiveness
2Measurement precision
If biochemical affinity measurements are used, then binding strength can be quantified, but endogenous processing and transport processes are neglected
Solution Approach 1:
The patent segments the complex multi-allelic HLA system into individual single-allele cell lines. This segmentation allows simultaneous study of endogenous processing and allele-specific binding without the confounding effects of multiple alleles, preserving complete information while enabling precise quantification
Solution Approach 2:
The patent uses mass spectrometry as an intermediary technique that can detect and quantify peptide-HLA complexes while also providing information about endogenous processing. This intermediary approach bridges the gap between binding quantification and processing information, capturing both aspects simultaneously
3Adaptability or versatility
If multi-allelic cell lines are used for MS-based approaches, then a broad portrait of peptide population is obtained, but allele-specific motif learning is complicated
Solution Approach 1:
The patent segments multi-allelic populations into individual single-allele cell lines, enabling clear allele-specific motif learning while maintaining the ability to study broad peptide populations. Each segmented system provides unambiguous allele-peptide associations, simplifying analysis complexity
Solution Approach 2:
The patent creates a universal platform of single-allele cell lines that can be used for both allele-specific motif learning and broad peptide population analysis. Each cell line serves multiple functions: studying its specific allele's preferences and contributing to overall peptide repertoire characterization, thereby resolving the complexity issue
4Quantity of substance
If large cellular input is required for LC-MS/MS methods, then sufficient peptide material is obtained for sequencing, but throughput is limited
Solution Approach 1:
The patent creates multiple copies of single-allele cell lines, each producing sufficient peptide material for MS analysis. This copying strategy eliminates the need to process large numbers of cells from primary samples, enabling high throughput while maintaining adequate peptide quantities for sequencing
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables the identification of a large number of HLA binding peptides specific to each allele, improving the accuracy of peptide prediction and overcoming the limitations of existing methods by considering endogenous processing and transport, thus enhancing the ability to predict immunogenic peptides for vaccine design and cancer therapy.
Implementation Method 1
Mass spectrometry (MS)-based approaches yield a large and relatively unbiased portrait of the population of processed and presented peptides
Data Source
AI summary
Adaptive immune responses rely on the ability of cytotoxic T cells to identify and eliminate cells displaying disease-specific antigens on human leukocyte antigen (HLA) class I molecules. Investigations into antigen processing and display have immense implications in human health, disease and therapy. To extend understanding of the rules governing antigen processing and presentation, immunopurified peptides from B cells, each expressing a single HLA class I allele, were profiled. A resource dataset containing thousands of peptides bound to distinct class I HLA-A, -B, and -C alleles was generated by implementing a novel allele-specific database search strategy. Applicants discovered new binding motifs, established the role of gene expression in peptide presentation and improved prediction of HLA-peptide binding by using these data to train machine-learning models. These streamlined experimental and analytic workflows enable direct identification and analysis of endogenously processed and presented antigens.