A complex trait genomics approach to investigating amyotrophic lateral sclerosis

Abstract

Amyotrophic lateral sclerosis (ALS) is the most common form of motor neuron disease (MND). It is a fatal neurodegenerative disease that has a lifetime risk of 1 in 300 people worldwide, with death often occurring within 3 to 5 years from the onset of symptoms. Currently, there is no effective treatment and no cure for ALS, in part due to a limited understanding of its aetiology. Most ALS genetics research to date is driven by discoveries in the familial form of the disease, where causality seems to reflect a heritable single gene mutation. Only a small proportion of ALS research is focussed on investigating the cause of sporadic ALS, which is likely to be a complex genetic disorder with many factors contributing to its aetiology, including genetic and environmental risk factors. The traditional ALS genetic research approaches which only rely on familial data or the single-trait Genome Wide Association Study (GWAS) paradigm might not be ideal for studying complex disorders like ALS. Therefore, the research presented in this thesis aims to study the aetiology of ALS using statistical techniques that have been developed for genetic analysis of complex traits applied to publicly available summary statistics level GWAS and multi-omics data. Newly available gut microbiome data was also used to understand non-genetic factors on the development of disease.Using more than 200 publicly available GWAS summary statistics from various traits to estimate genetic correlations with ALS, I found a negative correlation between ALS with cognitive performance (CP) and with education attainment (EA). This analysis also confirmed the previously reported positive correlation with schizophrenia (SCZ). Adding support for these genetic correlations, I was able to leverage ALS polygenic risk score prediction by multi-traits methods (MTAG and SMTpred) using these correlated traits. Furthermore, cell-type enrichment analyses of these traits showed that many central nervous system tissues relevant to ALS are also found to be highly significant in its correlated traits (CP, EA, SCZ). Genes expressed in dendritic cell were found to be significantly enriched in ALS association results which might suggest immune system involvement in ALS aetiology.The post-GWAS analysis of the largest unpublished ALS European ancestry GWAS (139,452 individuals) using Bayesian methodology suggested that ALS genetic architecture is likely to be less polygenic, has greater relative contribution from rare variants and has greater indication of negative selection compared to other common diseases. Using Summary Based Mendelian Randomisation (SMR) analysis, I provided support that ALS associated SNPs could play a causal role mediated through gene expression in brain and blood (SCFD1, RES18, MOBP, SLC98A, GGNBP2, DHRS11, ZNHIT3, MYO19, G2E2, SARM1). Moreover, most of these causal gene annotations are linked to protein processing in endoplasmic reticulum and cytoskeletal function that is relevant to previously reported ALS aetiology. By comparing SMR result with the correlated traits, I also found that CP has causal gene overlap with ALS. Moreover, the SMR significant genes for CP have an inflated SMR P-value distribution which could suggest functional overlap in both traits.The gut microbiome study of MND (where the majority of cases are ALS), suggests that there is no significant difference in term of composition or richness between ALS and healthy controls. The result showed that 4 ALS patients had dysbiosis, but with our current sample size, the possibility that this dysbiosis is a chance result cannot be excluded. By analysing the ALS patient survival status with the microbiome richness index (Shannon index), ALS patients with higher microbiome richness were found to have faster progression and mortality. This finding challenges the common notion that higher microbiome richness leads to better health. Despite all of these findings, it is still not clear whether the gut microbiome composition contributes to the cause of ALS or is just a consequence of ALS.