Concordance of next generation sequence-based and sequence specific oligonucleotide probe-based HLA-DRB1 genotyping

详细信息    查看全文

• 作者：Julie A. Lane ^{jlane@chori.org" class="auth_mail" title="E-mail the corresponding author} ; Jameel R. Johnson¹ ; ^{jameel.johnson@bcm.edu" class="auth_mail" title="E-mail the corresponding author} ; Janelle A. Noble ; ^{jnoble@chori.org" class="auth_mail" title="E-mail the corresponding author}
• 关键词：HLA ; Human Leukocyte Antigen ; NGS ; next generation sequencing ; SSOP ; sequence-specific oligonucleotide probe ; T1D ; type 1 diabetes ; PGM ; Personal Genome Machine ; MID ; multiplex identifier
• 刊名：Human Immunology
• 出版年：2015
• 出版时间：December 2015
• 年：2015
• 卷：76
• 期：12
• 页码：939-944
• 全文大小：495 K

文摘

Next generation sequencing (NGS) of clonally amplified DNA, using Roche 454 technology, was used to genotype HLA-DRB1, DRB3, DRB4, and DRB5 loci (exon 2 only) from a set of 993 samples from newborns with maternally-reported African American ancestry. DRB1 exon 2 was genotyped previously on the same sample set using sequence-specific oligonucleotide probe (SSOP) technology. Comparison of the genotype calls from both methods indicated concordance of 92.3%. Some discordance was expected due to the higher resolution of NGS data, compared to SSOP data. This resulted from selection of the incorrect allele from the ambiguity string produced by SSOP genotyping. Of 76 discordant genotypes, only three were due to resolution of ambiguity with the NGS method. The low percent of changes due to the increased resolution of the NGS method instills confidence in the overall value of previous data genotyped with moderate resolution methods, i.e., the vast majority of alleles present in a population are those that are detectable at moderate resolution. The remaining 73 discordant genotypes resulted from preventable errors in sample handling, data interpretation, and data entry. These results underscore the potential for error that can result from factors such as low quality genomic DNA, manual data entry, and interpretation of marginal genotyping results. Optimization of genomic DNA quality, automation of genotyping steps wherever possible, and use of the highest resolution technology available can lead to dramatic improvements in HLA genotype data quality. NGS-based methodology generated data of superior quality and accuracy compared to the SSOP system.