Integrating multimodal data to support Alzheimer’s disease target prioritization

Abstract

AbstractBackgroundThe limited success of therapeutics targeting the amyloid hypotheses of Alzheimer’s Disease (AD) suggests the need to examine additional and potentially complementary therapeutic approaches. Recent efforts to amass genome‐scale data from diverse cohorts have revealed thousands of new therapeutic targets and highlighted previously underappreciated molecular and cellular dysfunction associated with disease states. Our understanding of the mechanistic relevance of individual targets and the inter‐relationships of groups of targets remains limited. Here we combine evidence across multiple independent studies to prioritize specific targets for further resource and reagent generation in the Target Enablement to Accelerate Therapy Development for Alzheimer’s Disease (TREAT‐AD) consortium.MethodWe propose a gene‐level scoring method that synthesizes evidence from diverse studies and data modalities. Genetic scores are derived from genome‐wide association studies of large AD cohorts as well as genetic evidence from model systems. Multi‐omic scores combine evidence from transcriptomic and proteomic studies of human post‐mortem brain tissue in the Accelerating Medicine Partnership in Alzheimer’s Disease (AMP‐AD) cohorts. We also use scores based on high‐throughput functional screens in D. melanogaster and scores based on term associations from indexed publications. These subscores from each category are combined in a final target score that is used for target prioritization and hypothesis evaluation.ResultWe have scored over 17,500 targets using these scoring metrics including four pilot hypotheses implicated by recent work in the AMP‐AD consortium. The most significant functional annotations associated with high‐scoring targets include cellular respiration, regulation of trans‐synaptic signaling, neuron death, and activation of immune response. Using the highest‐priority targets within each pilot hypothesis, ten targets were further prioritized for immediate development of resources by the TREAT‐AD consortium.ConclusionThis prioritization provides an integrative and evidence‐based metric for the assessment of novel therapeutic targets and hypotheses. High‐priority targets will be considered for further assay development, structural biology, and medicinal chemistry studies in the TREAT‐AD consortium. These resources will be openly shared with the research community, which will help to de‐risk the evaluation of understudied candidate targets and therapeutic hypotheses.