Developing Novel Strategies for Personalized Treatment and Prevention of ALS: Leveraging the Global Exposome, Genome, Epigenome, Metabolome, and Inflammasome with Data Science in a Case/Control Cohort

Patients with amyotrophic lateral sclerosis (ALS), a rare fatal neurodegenerative disease, face a lengthy diagnostic process, and, although most survive only 2 to 4 years from diagnosis, lack information regarding their specific anticipated disease course due to a lack of prognostic tools. Although ALS is a heterogeneous disease of varied etiology, peripheral immune system dysfunction is ubiquitous, reflected by an altered whole blood transcriptome. Herein, we profiled whole blood gene expression by RNA sequencing in a large cohort of ALS cases versus controls. Several machine learning classifiers trained on our gene expression dataset predicted case-control status and survival, and integration analysis with external cohorts led to the identification of drug candidates. Data available via dbGaP include the de-identified subject- and sample-level data: raw and processed RNA-seq (BAM, count matrices) and key clinical phenotypes (diagnosis, demographics, survival). Of the 694 subjects in this study, 388 consented to share their individual-level data.]]>
Inclusion criteria: Age >18 years, can communicate in English. ALS cases must have an ALS diagnosis based on El Escorial criteria.]]>