Streaming Long-Read Sequence Alignments for HLA Predictions Using HLAminer

Area of Science:

• Genomics
• Immunogenetics
• Bioinformatics

Background:

• The human leukocyte antigen (HLA) gene cluster is crucial for immunity and highly polymorphic.
• Analyzing HLA regions traditionally requires high-resolution data and complex methods.
• Long-read sequencing offers potential for comprehensive HLA analysis.

Purpose of the Study:

• To present a streamlined protocol for predicting HLA alleles from long-read sequencing data.
• To demonstrate the efficiency and robustness of the HLAminer tool for HLA signature prediction.
• To enable accessible HLA typing from whole-genome shotgun data.

Main Methods:

• Directly streaming sequence alignments into HLAminer.
• Utilizing long-read sequencing data from Oxford Nanopore Technologies (ONT) and Pacific Biosciences (PacBio).
• Compatibility with any SAM file format-compatible read aligner, including minimap2.

Main Results:

• The protocol successfully predicts HLA class I and class II alleles.
• Robust HLA signature predictions were achieved even with older, lower-accuracy nanopore data.
• Effective analysis was demonstrated with as low as 10× sequencing coverage.

Conclusions:

• The presented method provides a simple, scalable, and efficient approach for HLA prediction from long-read sequencing data.
• This protocol democratizes HLA typing by leveraging readily available whole-genome shotgun data.
• The HLAminer tool facilitates accurate immune gene analysis across various sequencing datasets.