Genetic and genomic investigations of Amyotrophic lateral sclerosis

Amyotrophic lateral sclerosis (ALS) is a fatal, genetically heterogeneous neurodegenerative disease characterised by the loss of upper and lower motor neurons. Gene mutations remain the only proven cause of ALS. While 10% of patients have a family history (familial ALS; FALS), one third of these patients carry an unidentified causal mutation. Among the remaining 90% of apparently sporadic patients (sporadic ALS; SALS), less than 10% carry a known causal mutation. As such, a significant amount of genetic variation underlying ALS remains to be discovered. This thesis presents innovative approaches to identify novel genetic causes of ALS using next-generation sequencing (NGS). This involved the development and application of various bioinformatics strategies to whole-exome (WES) and whole-genome (WGS) sequencing datasets for various patient cohorts including FALS patients, families and ALS-discordant monozygotic twins. Assessment of the prevalence of known and candidate ALS genes among Australian patients revealed that 39.2% of FALS had an unidentified causal gene mutation, and identified eight candidate ALS mutations. Novel ALS gene discovery in four small families identified 19, 11, 16 and 64 candidate causal mutations in each. Having exhausted the genetic power of these families, an in silico pipeline was developed to assess the potential pathogenicity of each candidate mutation. This showed that five, six, one and 11 candidate mutations had a high potential to cause ALS. Gene discovery efforts in a fifth family using WES, WGS and genetic linkage data failed to identify any candidate mutations, however narrowed the search to just 14% of the genome. WGS of four ALS-discordant monozygotic twinsets also failed to identify any de novo mutations underlying disease discordance. This work expands our understanding of the genetic causes of ALS, and in turn provides much needed insight for the development of diagnostic and carrier-screening regimes, as well as relevant models of disease.