The landscape of genetic diseases in Saudi Arabia based on the first 1000 diagnostic panels and exomes

In this study, we report the experience of the only reference clinical next-generation sequencing lab in Saudi Arabia with the first 1000 families who span a wide-range of suspected Mendelian phenotypes. A total of 1019 tests were performed in the period of March 2016-December 2016 comprising 972 solo (index only), 14 duo (parents or affected siblings only), and 33 trio (index and parents). Multigene panels accounted for 672 tests, while whole exome sequencing (WES) represented the remaining 347 tests. Pathogenic or likely pathogenic variants that explain the clinical indications were identified in 34% (27% in panels and 43% in exomes), spanning 279 genes and including 165 novel variants. While recessive mutations dominated the landscape of solved cases (71% of mutations, and 97% of which are homozygous), a substantial minority (27%) were solved on the basis of dominant mutations. The highly consanguineous nature of the study population also facilitated homozygosity for many private mutations (only 32.5% of the recessive mutations are founder), as well as the first instances of recessive inheritance of previously assumed strictly dominant disorders (involving ITPR1, VAMP1, MCTP2, and TBP). Surprisingly, however, dual molecular diagnosis was only observed in 1.5% of cases. Finally, we have encountered candidate variants in 75 genes (ABHD6, ACY3, ADGRB2, ADGRG7, AGTPBP1, AHNAK2, AKAP6, ASB3, ATXN1L, C17orf62, CABP1, CCDC186, CCP110, CLSTN2, CNTN3, CNTN5, CTNNA2, CWC22, DMAP1, DMKN, DMXL1, DSCAM, DVL2, ECI1, EP400, EPB41L5, FBXL22, GAP43, GEMIN7, GIT1, GRIK4, GRSF1, GTRP1, HID1, IFNL1, KCNC4, LRRC52, MAP7D3, MCTP2, MED26, MPP7, MRPS35, MTDH, MTMR9, NECAP2, NPAT, NRAP, PAX7, PCNX, PLCH2, PLEKHF1, PTPN12, QKI, RILPL2, RIMKLA, RIMS2, RNF213, ROBO1, SEC16A, SIAH1, SIRT2, SLAIN2, SLC22A20, SMDT1, SRRT, SSTR1, ST20, SYT9, TSPAN6, UBR4, VAMP4, VPS36, WDR59, WDYHV1, and WHSC1) not previously linked to human phenotypes and these are presented to accelerate post-publication matchmaking. Two of these genes were independently mutated in more than one family with similar phenotypes, which substantiates their link to human disease (AKAP6 in intellectual disability and UBR4 in early dementia). If the novel candidate disease genes in this cohort are independently confirmed, the yield of WES will have increased to 83%, which suggests that most "negative" clinical exome tests are unsolved due to interpretation rather than technical limitations.