High-Resolution HLA Typing by Long-Read Sequencing — Unambiguous Full-Length Allele Resolution for Transplantation & Pharmacogenomics Research

CD Genomics provides unambiguous, full-length HLA allele resolution using PacBio Revio HiFi (>Q30) and ONT PromethION long-read sequencing. Unlike NGS short-read methods that rely on computational phasing of 150–300 bp fragments, our long-read platform spans complete 3–12 kb HLA genes in single reads, delivering definitive two-field and four-field allele typing for transplantation immunology, pharmacogenomics, and immuno-oncology research.

Human leukocyte antigen (HLA) genes constitute the most polymorphic region in the human genome, with over 38,000 known alleles documented in the IPD-IMGT/HLA database. Achieving unambiguous, allele-level HLA typing requires sequencing technology capable of resolving highly homologous, haplotype-phased, full-length gene sequences across the 4 Mb MHC region on chromosome 6p21.3. While most mainstream HLA typing services rely on next-generation sequencing (NGS) short reads (150–300 bp) that cannot span complete HLA genes — resulting in ambiguous phasing, unresolved cis-trans polymorphisms, and allele code reporting — CD Genomics takes a fundamentally different approach. We specialize exclusively in long-read sequencing-based HLA typing, leveraging PacBio Revio with HiFi read accuracy exceeding Q30 and Oxford Nanopore PromethION with ultra-long read lengths exceeding 100 kb to deliver unambiguous, full-gene phased allele resolution that NGS cannot achieve. This is not an add-on service — long-read sequencing is our core technology platform, purpose-built for applications where resolution beyond NGS capability is required.

Our service combines targeted or whole-genome long-read sequencing with dedicated bioinformatics workflows (HLAminer, ArcasHLA, or custom pipelines) to deliver unambiguous typing at two-field (four-digit) and four-field (eight-digit) resolution across all classical HLA class I (HLA-A, -B, -C) and class II (HLA-DR, -DQ, -DP) loci, as well as non-classical loci including HLA-E, -F, and -G. Every allele call is directly observed from reads spanning complete HLA gene sequences — not computationally inferred from short fragments as in NGS-based approaches.

Why Long-Read Sequencing for HLA Typing — Service Highlights

• Long-read is our core, not an add-on — Unlike general sequencing providers that offer long-read as an afterthought beside NGS, CD Genomics' HLA typing service is built entirely on long-read technology. We do not compromise with short-read hybrid approaches that reintroduce phasing ambiguity
• Dual-platform flexibility — PacBio Revio SMRT sequencing delivers high-accuracy (>Q30) HiFi reads, while ONT PromethION provides ultra-long contiguous reads spanning full HLA haplotypes
• Full-length allele resolution impossible with NGS — Complete gene-level phasing for all 12 classical HLA genes without computational inference or ambiguity codes. Short-read platforms cannot span the 3–12 kb HLA gene length in a single read, requiring probabilistic phasing that NGS services rely on
• Multiplexed throughput with rapid TAT — Up to 96 samples per Revio SMRT Cell 8M or PromethION flow cell with barcoded library preparation
• Flexible input compatibility — Targeted amplicon sequencing from whole blood, PBMCs, tissue, or genomic DNA (≥250 ng); also compatible with RNA-based typing for expression-level analysis
• End-to-end bioinformatics — HLAallele assignment, haplotype phasing, novel allele discovery, and optional expression quantification in a single deliverable package

Long-Read Sequencing Resolves HLA Alleles That Short-Read NGS Cannot Phase Unambiguously

The HLA region exhibits extreme polymorphism, with individual loci such as HLA-B harboring over 7,500 known alleles. This diversity underpins essential immune functions — antigen presentation, T-cell repertoire selection, and natural killer cell regulation — while simultaneously representing the primary genetic barrier to organ and hematopoietic stem cell transplantation (HSCT).

The NGS limitation that most HLA services share: Short-read next-generation sequencing (NGS) platforms, despite being the current mainstream methodology for HLA typing, produce reads of only 150–300 bp — far too short to span a full-length HLA gene (3–12 kb). This fundamental read-length constraint forces NGS-based HLA typing to rely on computational phasing and statistical inference to reconstruct alleles from fragmented data, resulting in ambiguous phasing, unresolved cis-trans polymorphisms, and reporting of allele codes (P-group or G-group) rather than definitive allele assignments. For applications where single-nucleotide resolution determines transplant compatibility or drug safety, these ambiguities are not merely inconvenient — they carry real consequences for research interpretation.

Why we choose long-read instead: Long-read sequencing overcomes these limitations at the source — by generating contiguous reads that span entire HLA genes in a single pass, eliminating the need for computational inference. The PacBio Revio platform produces HiFi reads of 15–25 kb with accuracy exceeding Q30 (>99.9%), enabling direct observation of phased haplotypes without statistical imputation. The ONT PromethION platform extends contiguity further, generating reads exceeding 100 kb that can span multiple HLA loci simultaneously, preserving long-range haplotype structure across the MHC region that no NGS platform can resolve. Studies have demonstrated that long-read HLA typing achieves four-field resolution accuracy exceeding 99% for class I and class II loci when benchmarked against reference materials.

Our Long-Read Platform Delivers Unambiguous Full-Gene Phasing at Four-Field Resolution

Bioinformatics Analysis

Sample Requirements for HLA Typing

Case Study: Targeted RNA-Based ONT Sequencing for High-Resolution Typing of 12 Classical HLA Genes

Johansson T, Koskela S, Yohannes DA, Partanen J, Saavalainen P. Targeted RNA-Based Oxford Nanopore Sequencing for Typing 12 Classical HLA Genes. Frontiers in Genetics. 2021;12:635601. (CC BY 4.0)

1. Background

HLA typing by short-read NGS faces persistent challenges due to the high polymorphism, mosaic structure, and long-range phase relationships across class I and class II genes. Johansson and colleagues developed a targeted RNA-based HLA typing method leveraging Oxford Nanopore single-molecule sequencing to evaluate whether full-length transcript sequencing could overcome these limitations and deliver unambiguous typing of all 12 classical HLA genes simultaneously.

2. Methods

Peripheral blood mononuclear cells from 50 healthy blood donors were used as the sample source. mRNA was reverse-transcribed into full-length cDNA using a template-switching oligo approach incorporating 10 bp molecular barcode. Twelve classical HLA genes (HLA-A, -B, -C, -DRA, -DRB1, -DRB3/4/5, -DQA1, -DQB1, -DPA1, -DPB1) were enriched by multiplexed PCR using locus-specific primers, divided into two gene pools, barcoded with ONT PCR barcoding kit, and sequenced on 10 MinION R9.4 SpotON flow cells on a MinION Mk 1b device. Basecalling was performed on the cloud-based Metrichor platform. HLA alleles were assigned using SeqPilot software (JSI Medical Systems) against the IPD-IMGT/HLA Database (release 3.27.0) and validated against Luminex SSO-PCR reference typing.

3. Results

Figure 2. HLA typing accuracy across 50 individuals by ONT RNA sequencing. (A) Allele calling rate at two-field resolution for class I versus class II loci. (B) Gene-level allele calling rates for all 12 classical HLA genes. Adapted from Johansson et al. (2021, Frontiers in Genetics, CC BY 4.0).

Key Findings

• >30,000 reads generated per individual across 10 multiplexed MinION runs, with approximately 17,000 high-quality aligned reads per sample contributing to allele assignment
• 99–100% accuracy at low-resolution (one-field) level for all 12 classical HLA loci, confirming the method's robustness for broad allele group typing
• 80% class I and 95% class II alleles called at two-field (four-digit) resolution or higher, with HLA-DRA, -DRB1, -DRB3, -DQA1, and -DPA1 achieving 98–100% call rates
• Successful multiplexing of 50 individuals across 10 flow cells (5 samples per flow cell) demonstrated the scalability of ONT RNA-based HLA typing for cohort studies

4. Significance

This study established that ONT RNA sequencing — performed on the MinION platform, the precursor to today's PromethION systems — can produce accurate high-resolution HLA typing across all 12 classical loci from a single experimental workflow. The method simultaneously provides full-length transcript sequences, enabling both genotyping and expression-level analysis from the same dataset. These findings directly support the application of contemporary ONT PromethION sequencing for high-resolution HLA typing at research scale, with substantially higher throughput and longer read capabilities than the MinION platform used in this foundational study.

When to Choose Long-Read HLA Typing — and When Alternative Methods May Suffice

Choose long-read HLA typing when:

• Full allele resolution is required for your study — Transplantation matching, pharmacogenomic screening, and immuno-oncology studies benefit from definitive four-field allele calls that short-read methods cannot provide
• Novel or rare allele discovery is a stated goal — Long reads capture complete gene sequences without probe bias, enabling detection of previously uncharacterized alleles
• Haplotype phasing across multiple HLA loci is needed — Ultra-long ONT reads can span adjacent HLA genes in a single read, preserving long-range phase information across the MHC region
• Your cohort requires consistent, high-resolution data — Multiplexed long-read sequencing delivers uniform coverage across 96 samples per flow cell with unambiguous allele calls for every sample

Alternative methods may be sufficient when:

• Low-resolution screening is acceptable — Population-level HLA frequency studies or preliminary association screens may not require four-field resolution
• Sample quality or quantity is limited — Severely degraded DNA (fragment size <5 kb) or input amounts below 100 ng may not yield sufficient long-read data; in such cases, targeted short-read methods may be more practical
• Only one or two specific alleles need confirmation — Targeted PCR-SSP or Sanger sequencing of specific exons may be more cost-effective for single-allele confirmation in small sample sets

CD Genomics provides free project consultation to help determine whether long-read HLA typing is the right approach for your specific research question. Contact our scientists to discuss your project requirements.

Sample Deliverables

1. Allele assignment report at two-field and four-field resolution for all 12 classical HLA genes

2. Phased haplotype assignments across the MHC region with read-level supporting evidence

3. Sequencing quality metrics including per-sample coverage depth, read length distribution, and error profile across HLA loci

4. Optional: novel allele discovery report with de novo consensus sequences for candidate novel alleles

5. Optional: allele-specific expression quantification for RNA-based typing projects