Improved detection and clinical interpretation of rare disease variants in complex genomic regions using nanopore sequencing
Reconstruction of complex genomic regions including recent segmental duplications, as well as the detection of repeat expansions and complex structural variants (SVs), have been challenging using short-read sequencing. These regions, however, potentially harbor a substantial fraction of the disease-causing variants missed in the large number of unsolved rare disease cases (still around 50% of all cases). The introduction of nanopore long-read whole-genome sequencing (LR-GS) into clinical genetic testing promises to reduce this gap by resolving complex genomic regions, facilitating haplotype phasing and by revealing complicated SVs that often consist closely located inversions, tandem-duplications and deletions.
Therefore, we have recently launched a study group consisting of four German university clinics aiming to evaluate the added diagnostic value of nanopore sequencing and to demonstrate its sustainability in clinical practice. Our study group relies on established collaborative structures with multidisciplinary teams (MDTs), data analysis task forces (DATFs), and data interpretation task forces (DITFs). In addition to a cohort of 1000 unsolved cases, we are currently analyzing a small cohort of solved cases (featuring causal repeat expansions, variants in duplicate genes, and complex rearrangements) and multiple GiaB reference samples (NA12878, NA12891, NA12892) to benchmark detection and clinical interpretation of variants for accreditation of LR-GS.
In this talk, I will focus on our experience with the bioinformatics analysis of nanopore LR-GS data for rare disease diagnostics, including development and/or application of tools and pipelines for SNV, indel and SV detection, haplotype phasing, estimation of repeat expansion length, genotyping in duplicate genes (e.g., GBA) and haplotype-specific DNA methylation analysis. I will discuss our quality criteria and benchmarking efforts and demonstrate the need for generating large Nanopore LR-GS background dataset to facilitate systemic filtering and efficient clinical interpretation of structural variants in clinical diagnostics.
