Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease characterized by progressive muscle paralysis due to selective loss of upper and lower motor neurons. ALS spreads quickly, typically leading to death from respiratory failure in 3-5 years. Currently, the genetic architecture of sporadic ALS is poorly understood. Some evidence suggests that a few rare genes, each with a large effect, contribute to ALS risk (oligogenic). In contrast, other evidence suggests that a large number of genes, each contribute a small amount to ALS risk (polygenic) and that many of these genes also play a role in other diseases (pleiotropy). A major obstacle to progress in the field has been that genome-wide association studies, the current state-of-the-art method for studying the genetic basis of disease, are statistically underpowered – able to detect only a subset of true genetic associations. For this reason, our group has been working to develop methods to boost statistical power for gene discovery in order to uncover the full genetic architecture of diseases like ALS.
Using these methods, we have found that ALS is a complex trait with a polygenic genetic architecture. Specifically, we have identified 96 ALS risk genes across 21 chromosomes conditioned on 65 distinct traits and diseases. We find that ALS has genetic overlap with several other diseases, including frontotemporal dementia, coronary artery disease, and major depressive disorder. We also find that a large number ALS risk genes are co-expressed together, show genetic interactions, and demonstrate physical interactions of their protein products. Within these genetic networks, some genes are more highly interconnected than others (e.g., ATXN2, SMARCA2, SCDF2, UGCG, COL16A) suggesting that they function as key ‘hubs’ in the genetic architecture of ALS and marking them as particularly important for further investigation. Of these genes we find that COL16A and UGCG are differentially expressed in brain tissue from ALS patients compared to controls. Collectively, our work has identified several novel genes that contribute to ALS risk and establishes ALS as a complex genetic trait with a polygenic architecture. These findings have important implications for ALS diagnosis and classification, in particular, they suggest that by focusing on the polygenic architecture of ALS we may be able to quantify an individual’s disease risk and stratify patients into more biologically valid subgroups for clinical trials.