Seedlabs
Source research

The IPD-IMGT/HLA database: recent developments in sequence submission

Dominic J. Barker, Richard H L Natarajan, Michael Cooper, Sebastian J. F. Hopper, Andrew D. Yates, Peter Parham et al. · 2025 · 156 citationsRead the paper

The IPD-IMGT/HLA Database, https://www.ebi.ac.uk/ipd/imgt/hla/, contains >43 000 unique alleles from 47 genes recognized by the WHO Nomenclature Committee for Factors of the HLA System. The HLA system, located within the human major histocompatibility complex on chromosome 6, represents the most polymorphic genes in the human genome and is described as hyperpolymorphic. With the rapid growth of HLA sequencing worldwide, traditional approaches for submitting and accessing data have been challenged. To address these challenges, the IPD-IMGT/HLA Database has expanded its infrastructure by developing Application Programming Interfaces, enabling flexible queries via both a web-based interface and programmatic access. A major advancement is the launch of a new HLA Submission Tool, replacing the previous system after 26 years and >28 000 submissions. The updated tool provides a streamlined user experience, extensive data validation, and automated submissions through an application program interface (API). Importantly, it eliminates the burden of duplicate submissions by offering optional brokering to the European Nucleotide Archive on our submitters' behalf, representing sequences in both databases while ensuring attribution. Together, the new submission tool, submission API, and ENA brokering reduce workload, increase efficiency, and provide ongoing support for the global HLA community.

What the full text shows

Deployed in practice
  • The database contains unique alleles from 47 genes. — >43,000 unique alleles (As of July 2025)
  • The new HLA Submission Tool reduced the average processing time for submissions. — 33% reduction (to 22 working days) (Compared to the prior 4-year period)
  • The Sequence Feature Annotation Tool (SFAT) achieved high accuracy in automated annotation of exon-intron boundaries and protein translation. — 99.7% accuracy (Validated on all submissions made to the database)
  • Increase in the number of alleles for non-classical HLA genes (HLA-E, -F, -G, -DM, -DO, MIC, and TAP). — 56% increase (Over the past three years)
  • Increase in the total number of alleles since the 2023 publication. — 23% increase (8,148 alleles added) (Between 2023 and July 2025)
  • Proportion of alleles with genomic sequences in the database increased. — From 51.2% to 57.6% (Since the 2023 publication)
Setup
A global database containing >43,000 alleles from 47 genes, with submissions from 51 countries and users in over 160 countries.
How
The researchers redeveloped the HLA Submission Tool with a new web interface and API, and implemented a brokering service to the European Nucleotide Archive (ENA). They integrated the Sequence Feature Annotation Tool (SFAT) for automated data validation and annotation.
Limitations
High proportion of partial sequences (42.4%) due to historical sequencing methods. · Difficulty in sequencing HLA class II genes (specifically HLA-DRB1) due to complex intronic regions. · Data mining of some large-scale projects (e.g., 1000K genomes) is prevented by INSDC deposition requirements and sequencing depth concerns.

Summarized by Gemma 4 from the open-access full text · 9 Oct 2026. Read it