Blood donor genotyping - how can long range sequencing help?

Nicholas Gleadall (University of Cambridge) first introduced the basics of blood transfusion. He noted that, as well as being widely used as a clinical intervention to treat blood loss, some individuals with rare diseases such as Sickle Cell anaemia are dependent on blood transfusions. To prevent alloimmunisation, it extremely important to correctly match the blood types of donor and recipient: Nicholas described how transfusion of only 10 ml of blood mismatched for ABO is enough to cause death. Nicholas briefly outlined the basis of antibody-based typing via serology, but noted that reagents are not available for all blood groups: 85% of donors have no common blood type data, aside from ABO and RH, and 94% of donors have no rare blood type data.

For this reason, Nicholas explains, "in recent years, blood banks are turning to DNA". He showed a Circos plot displaying blood group and platelet proteins, and the genes underlying them. He also showed the antigens identified for genotype versus serotype analysis: the results are incredibly accurate, but not perfect, and Nicholas stressed that as an incorrect identification could kill, even low error is not tolerable. He went on to describe his work showing how nanopore sequencing can be used to investigate these errors. He showed the very few coding positions that denote the differences between A, B and O blood types: these mutations were put into an algorithm, in which first O variants were searched for in the data, then B changes, then A. Instances resulting in discrepancies after this algorithm were selected for whole genome sequencing on the PromethION: 15x coverage was generated from a single flow cell. In the example given, the data revealed a crossover of group O variants into a B background - a rare event. Prior to sequencing, the team had suspected a structural variation caused the discrepancy.

Nicholas then moved on to a case study of a patient recruited to the NIHR Rare Disease project, with a bleeding disorder (VWF), who was in need of transfusions. Serological analysis indicated that she lacked all Rh system antigens: Nicholas described how her blood type is "ultra-rare", and has only been identified in 43 people globally. A sample was sent to global reference labs, and whole genome sequenced; through this, a splice site variant was identified in the RHAG gene - however, this was heterozygous, so didn't explain the phenotype observed. Nicholas, having been told "you're not passing your viva unless you solve this case", nanopore sequenced the sample, using sniffles for structural variant calling and WhatsHap for haplotyping. Along with confirming the splice site variant, he identified a 1,957 bp tandem duplication in the RHAG gene, which was subsequently confirmed via Sanger sequencing. Nicholas highlighted this as a good example of what nanopore sequencing can do for research into rare variants.

Nicholas described how the NIHR Rare Disease project has 30x short-read WGS data for over 13,000 patients; nanopore sequencing on the PromethION to 15x coverage will be used to investigate ~100 unresolved cases. Nanopore sequencing has already identified Glanzmann's thrombaesthenia, through an inversion in exon 9, in one sample (NIHR BioResource, Nature - in review). Nicholas stated that "nanopore sequencing is excellent for resolving complex genomics", and that high-quality haplotyped reference sequences "redefine knowledge of structural variants" such as in blood group genes. Finally, he described how the data will be used to improve algorithms for genomic analysis.

Authors: Nicholas Gleadall