The utility of ultra-deep RNA sequencing in Mendelian disorder diagnostics.

RNA sequencing (RNA-seq) has emerged as a powerful tool for resolving variants of uncertain significance (VUSs), particularly those affecting gene expression and splicing. However, most reference datasets and diagnostic protocols employ relatively modest sequencing depths (∼50-150 million reads), which may fail to detect low-abundance transcripts and rare splicing events critical for accurate diagnosis. We evaluated the diagnostic and translational utility of ultra-high-depth (up to ∼1 billion unique reads) RNA-seq in four clinically accessible tissues using the Ultima sequencing platform. After validating the performance of Ultima RNA-seq, we investigated how increasing sequencing depth affects gene and isoform detection, splicing variant discovery, and clinical interpretation of VUSs. Deep RNA-seq substantially improved sensitivity for detecting lowly expressed genes and isoforms, achieving near saturation for detection at 1 billion reads. In two probands with VUSs, pathogenic splicing abnormalities were undetectable at 50 million reads but emerged at 200 million reads, becoming even more pronounced at 1 billion reads. Using deep RNA-seq data, we constructed a resource, MRSD-deep, to estimate the minimum required sequencing depth to achieve desired coverage thresholds. MRSD-deep provided gene- and junction-level guidelines, helping labs select appropriate coverage targets for specific applications. Leveraging deep RNA-seq data on fibroblasts, we also built an expanded splicing-variation reference that successfully identified low-abundance splicing events missed by standard-depth data. Our findings underscore the diagnostic and research benefits of deep RNA-seq for Mendelian disease investigations.