Fig. 1: Outline of the case-control study design.

A case-control analysis was performed to gain insights into the role of genotypes involving dual heterozygosity for TYR:c.1205 G > A (p.Arg402Gln) [rs1126809] and OCA2:c.1327 G > A (p.Val443Ile) [rs74653330] in albinism. These two missense changes were selected as they are the commonest albinism-related variants in European-like populations. The majority of participants in the “case” cohort (1015/1120) were identified through the database of the University Hospital of Bordeaux Molecular Genetics Laboratory, France. All these probands had at least one key ocular feature of albinism, i.e. nystagmus or prominent foveal hypoplasia (see “Methods”). The remaining 105/1120 cases were identified through the Genomics England 100,000 Genomes Project dataset and had a diagnosis of albinism or partial/ocular albinism. The “control” cohort included 29,451 unrelated individuals from the Genomics England 100,000 Genomes Project dataset, none of whom had a recorded diagnosis of albinism. To reduce the likelihood of obtaining spurious signals due to population stratification effects or due to the presence of albinism-related variants other than the two studied changes, we focused only on individuals who: (i) were projected to have European-like ancestries; (ii) did not have a genotype in keeping with a molecular diagnosis of albinism; (iii) were not heterozygous for a pathogenic or a likely pathogenic variant in TYR or OCA2 (with the exception of TYR:c.1205 G > A and OCA2:c.1327 G > A). The following two covariates were used: sex and number of common albinism-associated alleles, i.e. DNA sequence alterations in the genomic locations corresponding to TYR:c.− 301 C > T [rs4547091] and TYR:c.575 C > A (p.Ser192Tyr) [rs1042602]. These two variants have been previously shown to modify the effect of TYR:c.1205 G > A, a common missense change that can act as a low penetrance variant^19,20,21. To assess the robustness of the findings we performed additional analyses in subsets of the cohort. Validation studies in an independent cohort (UK Biobank) were also conducted. 100K_GP, Genomics England 100,000 Genomes Project; het, heterozygous; UKB, UK Biobank. TYR and OCA2 variant numbering was based on the transcripts with the following identifiers: NM_000372.5/ENST00000263321.6. and NM_000275.3/ENST00000354638.8.