Diagnostic yield and limitations of whole-genome sequencing for hereditary cerebellar ataxia

Less than half of the individuals with hereditary cerebellar ataxia receives a genetic diagnosis. Repeat expansions account for disproportionate number of hereditary cerebellar ataxia and have genetically heterogeneous causes. These genetic loci include ATXN1, ATXN2, ATXN3, CACNA1A, ATXN7, ATXN8OS, ATXN10, PPP2R2B, TBP, ATN1, FMR1, BEAN1, NOP56, GLS, THAP11, GAA-FGF14, ZFHX3, FXN and RFC1. This study aims to assess the yield of short-read whole genome sequencing in the molecular diagnosis of hereditary cerebellar ataxia. We recruited 380 patients (351 probands) from a national ataxia centre in United Kingdom. They underwent short-read whole genome sequencing as a part of the 100 000 Genomes Project. Bioinformatic pipeline of whole genome sequencing include variant prioritization in selected virtual gene panels, customized analysis with a focus on repeat expansions, structural variants and recently reported hereditary cerebellar ataxia genes. All potential genetic variants were reviewed in a multidisciplinary team, and further confirmation tests were performed as appropriate. Whole genome sequencing identified causative variants in 115 (33%) out of 351 probands. We established 46 distinct presumptive molecular diagnoses with the most frequent being SPG7 (n = 22), RFC1 (n = 20) and CACNA1A (n = 10). However, it failed to detect any probands with novel ataxia gene GAA-FGF14, which was subsequently identified on polymerase chain reaction screening in 10 unsolved probands. In conclusion, whole genome sequencing is a useful diagnostic test in hereditary cerebellar ataxia patients and can be used to detect repeat expansions, structural and mitochondrial variants. However, identification of complex structural variants and sizing of large repeat expansions remains a challenge and require alternative molecular testing techniques.