ABSTRACT
Single-molecule sequencing should allow for unambiguous, accurate, and high-throughput HLA typing. In this proof of principle study, we investigated the effects of fragment size for library preparation, indexing strategy, and read length on HLA typing. Whole gene amplicons of HLA-A, B, C, DRB1, and DQB1 were obtained by long-range PCR. For library preparation, two fragment sizes were evaluated: 100-300bp and 300-600bp. For sample multiplexing, two indexing strategies were compared: indexing-by-amplicon, where each individual amplicon is barcoded, and indexing-by-patient, where each patient's five loci are equimolarly pooled after PCR and indexed with the same barcode. Sequencing was performed on an Illumina MiSeq instrument using paired-end 150bp and 250bp read lengths. Our results revealed that the 300-600bp fragments in the 2×250 MiSeq group gave the most accurate sequencing results. There was no difference in HLA typing results between the two indexing strategies, suggesting that indexing-by-patient, which is much simpler, is a viable option. In conclusion, enzymatic fragmentation of pooled whole gene amplicons is a suitable strategy for HLA typing by next-generation sequencing on the Illumina MiSeq.