HLA Typing
✓ 已验证Dual-tool cross-validation · 4-digit resolution · Foundation of neoantigen pipeline
HLA typing is the first step of the tumor neoantigen pipeline--without accurate HLA typing, all downstream MHC binding predictions are built on sand. DiVo Gen²AI uses OptiType + Polysolver dual-algorithm cross-validation, not a single tool run. This page serves three audiences: patients and public wanting to understand HLA, hospitals and pharma seeking computational service partners, and investors and peers evaluating technical capabilities.
Read "What is HLA Typing" below--understand why it's the first step of immunotherapy.
Focus on "DiVo's Role", "4-Step Pipeline", "Benchmarks"--dual-tool cross-validation, not single run.
Focus on "Differentiation", "Benchmarks", "Glossary"--evaluate the technical barrier of this entry service.
HLA typing is the entry step for the following flagship services
What is HLA Typing
HLA (Human Leukocyte Antigen) is the immune system's "identity tag"--everyone's HLA type is different, determining how your immune system distinguishes "self" from "invaders". In tumor immunotherapy, HLA type determines which mutation fragments can be "seen" by your immune system.
HLA typing determines which HLA alleles a patient carries. This is the first step of the neoantigen pipeline--without accurate HLA typing, downstream MHC binding prediction, pMHC structure prediction, and immunogenicity scoring are all impossible. Wrong typing means the entire pipeline runs in vain.
DiVo doesn't use a single tool and accept the result, but uses OptiType and Polysolver two independent algorithms for cross-validation--two algorithms from different principles give consistent judgment, improving reliability by orders of magnitude.
HLA is the "display model" for immune presentation. Different models display different peptide fragments--without knowing the model, you can't predict which mutation peptides can be presented on the cell surface for T cell recognition.
A single algorithm may produce false positives/negatives due to sequencing quality or alignment bias. Two independent algorithms cross-validate--consistent results are trustworthy--this is the fundamental difference from "run once and deliver".
DiVo Gen²AI's Role
HLA typing is Step 1 of DiVo's neoantigen 8-step pipeline, and CT1 entry of the CAR-T 5-step pipeline. We provide end-to-end computing from sequencing data to final HLA typing results--not just running one tool for a report, but dual-tool cross-validation + consistency determination, ensuring typing results can directly serve as reliable input for downstream pipelines.
We do not do clinical HLA matching (for organ transplant), do not do clinical diagnostic interpretation of typing results. We deliver 4-digit HLA-I typing results that can be integrated into the neoantigen pipeline.
Core Capability · 4-Step Pipeline
From sequencing data to cross-validated HLA typing results
Sequencing Data QC
✓ 已验证FastQC + Trimmomatic
Clean reads
OptiType HLA-I Typing
✓ 已验证OptiType 1.5.0
4-digit HLA-I typing results
Polysolver Cross-Validation
✓ 已验证Polysolver v4
Independent HLA-I typing results + consistency report
Dual-Tool Consistency Determination
✓ 已验证DiVo in-house comparison logic
Final HLA typing + confidence annotation
Differentiation
Core differences from single-tool HLA typing
Dual-Tool Cross-Validation
OptiType + Polysolver run independently from different principles, cross-validation improves reliability. Single-tool typing may have false positives/negatives; dual-tool consensus required--in the neoantigen pipeline, wrong HLA typing invalidates the entire downstream pipeline.
4-Digit Resolution
Not coarse 2-digit typing (e.g., A*02), but precise to 4 digits (e.g., A*02:01). 4-digit resolution is a required input for MHC binding prediction--2-digit typing cannot distinguish subtypes with vastly different affinities within the same supertype.
Dual-Tool Cross-Validation
HLA-I Typing (4-digit resolution)
RNA-seq/WES-based rapid HLA typing
A*24:02, B*39:01
HLA-I Typing (4-digit resolution) · Cross-validation
WES-based independent HLA typing algorithm
A*01:01:01:01, B*07:02:01, C*01:02:01
Benchmarks
| Metric | Value | Note |
|---|---|---|
| HLA-I typing resolution | 4-digit | A*24:02, B*39:01 level |
| Dual-tool consistency | Passed | OptiType + Polysolver cross-validation |
| Typing methods | 2 | RNA-seq/WES dual algorithms independently |
| Pipeline position | Step 1 | Neoantigen 8-step pipeline entry |
Honest Boundaries
What we can and cannot do, clearly stated
What We Can Do
What We Don't Do
Glossary
5 core terms in HLA typing
| Abbr. | Full Name | Translation | Explanation |
|---|---|---|---|
| HLA | Human Leukocyte Antigen | Human Leukocyte Antigen | Human MHC genes, determining how the immune system distinguishes "self" from "non-self" |
| MHC | Major Histocompatibility Complex | Major Histocompatibility Complex | The "display board" on cell surfaces, presenting intracellular protein fragments to T cells |
| 4 位分辨率 | 4-digit resolution | 4-digit Resolution | e.g., A*02:01, precise to allele protein level, required input for MHC binding prediction |
| WES | Whole Exome Sequencing | Whole Exome Sequencing | Sequencing only the protein-coding regions (exons) of the genome, cost-effective |
| RNA-seq | RNA Sequencing | RNA Sequencing | Sequencing genes being expressed in cells, reflecting actual HLA gene transcription levels |
CAPACITY TRACE
能力回溯
这项服务由哪些能力支撑——从硅片到你的场景
硅片(L1) → 模型(L2) → Agent(L3) → 管线(L4) → 你的场景