Debarshi Mustafi from the University of Washington presented at the Nanopore Community Meeting in Houston on “Complex phased variants in inherited retinal diseases with long-read sequencing.” Mustafi talked about the thin layer of tissue forming the retina. They are interested in inherited retinal diseases (IRDs). For a long time, people thought these were auto-immunity-related diseases. Mustafi explained that there are now gene therapies and CRISPR Cas9 correction options. However, Mustafi emphasized that the genetics of IRDs are complex. Understanding the variants that are driving disease is challenging, yet Mustafi and others think the future of IRD care needs disease variant identification. The panels used for sequencing and genetics are very different in coding and non-coding genes. Thus, Mustafi said you may get a non-diagnostic result based on the selected panel. Mustafi described how long-read sequencing can help “phase” and detect responsible genes. The USH2A gene is 800,000 bp in size! Now, with adaptive sampling, you can sequence everything and enrich for this gene. Mustafi’s lab uses MinION devices and a computer with two NVIDIA RTX A6000 cards. With adaptive sampling, targeted long-read can produce enough data to do haplotyping and variant detection. The traditional method of phasing is through familial segregation and sequencing parents. With data from one flow cell, Mustafi and the team are able to identify haplotyped variants. Additionally, complex variants can be examined. Mustafi thinks the future potential of long-read sequencing in diagnostics will be rapid phased results. The team was able to identify variants with phasing with eleven hours of sequencing. In pediatric cases of retinoblastoma, rapid characterization of the RB1 gene could dictate care. Mustafi concluded that the sufficient accuracy from the MinION makes this work more accessible.
