(P-IG-12) Comprehensive Genome Variation Database Analysis for the CD59 Gene

The CD59 blood group system is based on the CD59 glycoprotein that is highly expressed in the membranes of many cell types. It protects the cells from complement-mediated cell lysis by inhibiting the assembly of the complement membrane attack complex. Lack of a functional CD59 protein has been shown to cause chronic hemolysis and immune-related neuropathies. Recent large-scale sequencing projects are generating datasets of human variation across genetically heterogeneous populations. Worldwide differences in the prevalence and distribution of genetic variants can affect a multitude of diseases. Understanding global genetic diversity can provide insights into the mechanisms of underlying disease and help develop public health decision-making models.

Variants in the CD59 gene have not been collated and compared for their pathophysiologic effects. We aimed at a comprehensive analysis of CD59 gene from publicly available databases to identify population-specific variants. We explored if computational analysis could discriminate deleterious from neutral variants.

Study

Design/Methods:

Variants in the CD59 coding sequence (exons 4, 5, and 6) and the splice sites were systematically compiled from 4 major populations in 6 whole genome and whole exome databases. The populations represented were African, Caucasian, Asian, and Latin American. PredictSNP metaserver was used to determine the functional impact of non-synonymous single nucleotide variants (SNVs).

Results/Findings:

Among 488,592 individuals analyzed, nucleotide variants were observed in only 6,491 individuals (1.33%). Most of the variants with >10 observations were synonymous and occurred in all 4 major populations analyzed. Among the non-synonymous variants, 9 had been deposited in ClinVar and 3 had been observed in patients and published. PredictSNP’s deleterious classification of 5 out of 9 SNVs correlated well with the diseased clinical outcome in ClinVar. The most common computationally predicted deleterious variant was the non-synonymous p.Pro128Ser present in the 1 in 11,187 Caucasians and 1 in 19,685 Latin Americans, yet has not been reported in a patient.

Conclusions: We collated an extended list of CD59 variants from genome databases, classified as neutral or deleterious based on PredictSNP. Our method can be applied to identify individuals with possible latent CD59 deficiency symptoms, such as hemolytic transfusion reactions. In conjunction with clinical data, CD59 variants could eventually guide personalized clinical decisions.