Autosomal recessive disorders represent a substantial burden of rare genetic disease, particularly in populations with high rates of consanguinity such as Saudi Arabia and Pakistan. Despite widespread use of next-generation sequencing (NGS), a significant proportion of affected consanguineous families remain genetically unsolved following standard clinical analysis. This project aimed to develop and apply integrated bioinformatic pipelines to identify pathogenic variants in previously unsolved recessive consanguineous families using short-read whole exome sequencing (WES), whole genome sequencing (WGS), long-read Oxford Nanopore sequencing, and phenotype-driven reanalysis.
As part of this study, multiple consanguineous families of diverse ancestry were recruited or re-analysed following unsuccessful clinical diagnostic testing. Standardised short-read analysis was carried out using homozygosity mapping, variant calling, and prioritisation with in silico tools including SpliceAI, AlphaMissense, and depth-of-coverage analysis. Candidate variants were assessed based on segregation, population frequency, and biological relevance.
A homozygous splice-site variant in COMMD4 was identified in a consanguineous family with a phenotype consistent with Ritscher–Schinzel syndrome, establishing COMMD4 as a novel disease gene. Biallelic variants in VPS36 were identified in a Saudi family (Family 133836) presenting with a severe progressive neurodevelopmental disorder with global cerebral atrophy, uncontrolled epilepsy, and progressive dystonia. Additional affected families with biallelic VPS36 variants were subsequently identified through international collaboration, establishing VPS36 as a novel neurodevelopmental disease gene.
For families unsolved by short-read analysis, Oxford Nanopore PromethION long-read sequencing was performed using the EPI2ME wf-human-variation workflow, integrating small variant calling (Clair3), structural variant detection (Sniffles2), and copy number analysis (Spectre). This approach confirmed a PARS2 missense variant as the cause of developmental and epileptic encephalopathy-75 by excluding structural rearrangements and resolved a complex deletion within the opsin gene cluster that was undetectable by short-read sequencing, demonstrating the diagnostic value of long-read technology at repetitive and structurally complex loci.
Finally, Exomiser-based phenotype-driven reanalysis of unsolved cases from the 100,000 Genomes Project identified additional diagnoses in KPNA3, COL9A2, VPS13B, and HID1. This study established new genotype–phenotype correlations in COMMD4, VPS36, and KPNA3 that will be beneficial for future clinical diagnosis and patient care, and supports a tiered diagnostic strategy combining short-read sequencing, targeted long-read sequencing, and iterative phenotype-driven reanalysis for unsolved recessive consanguineous families.
