Quantifying prior probabilities for disease-causing variants reveals the top genetic contributors in inborn errors of immunity

We present a framework to quantify the prior probability of observing known disease-causing variants across all genes and inheritance modes. First, we computed genome-wide occurrence probabilities by integrating population allele frequencies, variant classifications, and Hardy-Weinberg expectations under autosomal dominant, recessive, and X-linked inheritance. Second, both pathogenic variants and missing causal candidates were tested to identify the most likely genetic disease determinant and provide a clear confidence range for the overall diagnosis. This provided a complete and interpretable summary of evidence for genetic diagnosis. Third, we summarised variant probabilities for 557 genes responsible for inborn errors of immunity (IEI), now integrated into a public database. Fourth, we derived new data-driven IEI classifications using protein-protein interactions and curated clinical features, aligned to immunophenotypes. Finally, we validated the framework in national-scale cohorts, showing close concordance with observed case numbers. The resulting datasets supported causal variant interpretation and evidence-aware decision-making in clinical genetics. ¹