“Long-read sequencing resolves cryptic structural variation in individuals with syndromic intellectual disability” was the title of the presentation Griet De Clercq from Ghent University in Belgium gave at the Nanopore Community Meeting 2022. De Clercq noted that “40% of people with intellectual disability (ID) remain without genetic diagnosis” and that current diagnostic tests used in hospitals don’t detect structural variations. The research team hypothesized that long-read technology could be used to identify de novo structural variants in their samples. The study they designed included 13 DNA samples from fresh blood “proband-parent trios.” Their wet-lab protocol used a standard Oxford Nanopore Technologies (ONT) procedure with size-selection, library prep with the LSK-109 kit, and sequencing on R9 flow cells. They used one PromethION flow cell for every individual! The sequencing produced an N50 of 20.6 kb with mean coverage of 25.9. For the bioinformatics, the team used guppy super accuracy model for base calling, filtered with nanofit, and aligned with minimap2. Structural variants were determined with sniffles2 and explored with causality. This approach resulted in 21,000 structural variants identified per individual, with the majority being deletions and insertions. While nearly 50% of all variants affect a gene, De Clercq noted that only a “small minority affects an exon.” Most of the calls are in repetitive regions and incorrect. Thus, they found one true de novo event in five samples. They concluded that zero to 1 true de novo event was found in their samples. The research team then analyzed a single sample with clinical diagnosis of Mowat-Wilson using a MinION flow cell. Nanopore long-read sequencing revealed the expected implicated gene, but it also uncovered complex rearrangement. This clinical example highlights the potential of this workflow and application of long-read sequencing.
