Researchers at Oxford Nanopore Technologies have developed a scalable sequencing workflow that could simplify the diagnosis of inherited disorders. Here, lead researcher Philip Rescheneder, shares findings that suggest long-read sequencing could improve the characterization of complex repeat expansion disorders while simplifying laboratory workflows, as presented at the AMP Annual Meeting & Expo 2026.
The following transcript has been edited for clarity.
Hello everyone. My name is Philip Rescheneder. I am Senior Director of Applications Bioinformatics at Oxford Nanopore Technologies, and I'm going to present a scalable research workflow for analyzing short tandem repeat expansions using whole genome sequencing or alternatively what we call digital panels.
Repeat expansions are associated with more than 60 inherited disorders, but routine testing is usually loci specific and doesn't allow for the characterization of the full breadth of variation in these loci.
Native long reads, on the other hand, span the full repeat, allowing us to not only investigate the length but also the repeat motive, any interruptions within the repeat, and even methodation at the same time.
For this work, we used a software-based targeting method called adaptive sampling which lets you define target regions in software and thus sequence a custom panel of repeat loci without needing any local specific wet-lab targeting.
We used a new analysis tool called “mahi mahi” to genotype repeats for both the whole genome sequencing data and data from our digital panels. We tested this whole workflow on four different groups of samples, checking how closely it matched other computational approaches and also orthogonal test data.
In the largest comparison against reference data from the thousand genomes project, our results agreed with an orthogonal assembly-based analysis method for about 99 percent of the nearly 11,000 alleles we looked at, with most of the remaining differences being explained by errors in the assembly approach.
When comparing our two sequencing approaches, whole genome sequencing and the digital panels in a set of well-studied reference samples, they agreed on repeat length for more than 99 percent of the tested alleles. We also correctly picked up every known expansion in 14 well-characterized reference samples.
Finally, when we reran archived samples from three independent laboratories using our workflow, our results matched the original result in 177 of the 180 cases we looked at. The three exceptions were only off by one or two repeat units and fell right at a category boundary.
Beyond just measuring repeat length, long reads allowed us to resolve finer details in this loci. For example, we identified interruptions within the repeat at the ATX and bond locus which were then confirmed by orthogonal restriction digestion methods and we also profiled a specific methylation in FM1.
Taken together, these results show that our whole genome sequencing and digital panels workflow provides a practical way to fully characterize tandem repeat loci at scale without the need for wet-lab-based targeting methods.
