Suzanne de Bruijn

242 Chapter 5 UCNEbase 64 and CraniofacialAtlas 65 . For each regulatory element, a gene interaction score (>7) and element confidence score (>0.7) are provided. The EnhancerAtlas V2 is a database providing enhancer annotations in different species based on experimental datasets determined in several tissues and cell types. WGS data were analyzed for variants located within these elements and two rare potentially regulatory variants (Chr7:107220628C>A, Chr7:107384987C>G) were identified in two M1 individuals (SLC002 and SLC045) ( Table S6 ). Both variants are located in a predicted enhancer element of SLC26A4 according to GeneHancer. We did not find any strong indication of a functional effect for the two variants based on (nucleotide) conservation scores (PhyloP, UCSC genome browser 44 ) or loss of transcription factor binding sites (JASPAR database 66 ). Therefore, the variants were considered non-pathogenic, although only a reporter assay can completely exclude a potential regulatory effect of the variants on SLC26A4 expression. A FOXI1 missense variant is revealed in three unrelated index cases Several studies have suggested a potential digenic inheritance for SLC26A4 variants and variants in KCNJ10 and FOXI1 . 67-69 Additionally, a more recent study suggested digenic inheritance with pathogenic variants in EPHA2. 70 We screened all remaining genetically unexplained individuals (M1, M0/CEVA and M0) for variants in these genes with an AF ≤5% (gnomAD V2.1.1). In cases for which only MIP sequencing data was available, coding regions and exon-intron boundaries of FOXI1 and the regions harboring the reported pathogenic variants in EPHA2 (c.1063G>A; p.(G355R), c.1532C>T; (p.T511M), NM004431.4) were analyzed using Sanger sequencing. In three individuals (SLC039; M0/CEVA, SLC052; M0 and SLC069; M0) a c.677C>T (p.(Thr226Ile)) FOXI1 (NM_012188.4) missense variant was identified ( Table 3 ).