Bioinformatics analyses of variants
To characterize the population frequency and the pathogenicity of the variants, we predicted the population allele frequency of the variants using 1000 genomes (https://www.internationalgenome.org/), gnomAD (https://gnomad.broadinstitute.org/) and ExAc allele frequency (http://exac.broadinstitute.org) based on the previous mentioned (Gudmundsson et al., 2021; Kobayashi et al., 2017; Zheng-Bradley & Flicek, 2017). dbSNP (https://www.ncbi.nlm.nih.gov/snp/), Clinvar (https://www. ncbi.nlm.nih.gov/clinvar/) and HGMD (Human Gene Mutation Database, http://www. hgmd.cf.ac.uk/ac/index.php) were used to predict the pathogenicity of the variants (Landrum et al., 2018; Sherry et al., 2001; Stenson et al., 2014). Pathogenicity of both variants was analyzed according to the American College of Medical Genetics and Genomics guidelines (Richards et al., 2015). MetaDome (https://stuart.radboudumc.nl/metadome/) (Wiel et al., 2019) combined with NCBI-BLAST (https:// blast.ncbi.nlm.nih.gov/Blast.cgi) to analyze the variant tolerance of TMEM126B exon 2 as well as exon 3 and amino acid conservation among mammals. Alternative splice site predictor (http://wangcomputing.com/assp/) (Wang & Marín, 2006) and exonic splicing regulation (http://krainer01.cshl.edu/tools/ESE2/) (Cartegni et al., 2003) to predict the effect of two variants onTMEM126B mRNA splicing.