Summary
G6PD deficiency, an enzymopathy affecting 7% of the world population, is caused by over 160 identified amino acid variants in glucose-6-phosphate dehydrogenase (G6PD). The clinical presentation of G6PD deficiency is diverse, likely due to the broad distribution of variants across the protein and the potential for multidimensional biochemical effects. In this study, we use bioinformatic and biochemical analyses to interpret the relationship between G6PD variants and their clinical phenotype. Using structural information and statistical analyses of known G6PD variants, we predict the molecular phenotype of five uncharacterized variants from a reference population database. Through multidimensional analysis of biochemical data, we demonstrate that the clinical phenotypes of G6PD variants are largely determined by a trade-off between protein stability and catalytic activity. This work expands the current understanding of the biochemical underpinnings of G6PD variant pathogenicity, and suggests a promising avenue for correcting G6PD deficiency by targeting essential structural features of G6PD.