Overcoming the pitfalls of NGS-based molecular diagnosis of Shwachman-Diamond syndrome

Purpose: Shwachman-Diamond syndrome (SDS) is predominately caused by biallelic mutations in the SBDS gene and is characterized by exocrine pancreatic insufficiency, skeletal abnormalities and pancytopenia. Gene conversion between SBDS and its pseudogene SBDSP1 is the major cause. We established an efficient approach, HapICE, to infer the haplotype of SBDS based on short-read next-generation sequencing (NGS). Methods: HapICE based on the Expectation-Maximization algorithm was developed to detect variants in exon 2 of SBDS arising from gene conversion. We retrospectively analyzed two common pathogenic variants (c.183_184delinsCT and c.258+2T>C) in suspected SDS patients and compared the results with those from conventional NGS analysis. Results: In 47 SDS high-risk individuals and 64 available parents, HapICE improved the diagnostic rate by 27.7% compared with conventional methods and revealed 100% (95% CI: 92.5%-100%) concordance with Sanger sequencing. In addition to eighteen patients having consistent genetic results by both methods, HapICE further reported 8 patients with more accurate variant detection and 13 cases with the c.183_184delinsCT variant missing by conventional methods. HapICE also showed better performance in screening for carrier and wild-type status. Conclusion: We have developed a novel SBDS variant detection tool through regular NGS data that demonstrated precise variant detection performance in clinical scenarios.