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

摘要 遗传遗传性不能完全解释肌萎缩侧索硬化症(ALS)，以及ALS基因突变的速度 发现已经放缓，这意味着需要全新的研究方向来揭开疾病的机制 并确定治疗方法。我们的目标是了解、治疗和预防肌萎缩侧索硬化。我们的总体方法是确定 ALS的暴露、基因组学、表观基因组学、转录组学、代谢组学和炎症的交叉。 我们的理论基础是，即使是基于有限的粗血浆测量的先前环境风险评分(ERS) 污染物的种类与肌萎缩侧索硬化症风险增加7倍、存活率减少2倍有关，因此 对暴露组和其他组学的详细了解可以立即提供新的、迫切需要的策略- 肌萎缩侧索硬化症的治疗和预防。我们提出了三个目标：1)综合评价环境 肌萎缩侧索硬化症与对照组的暴露和多基因因素以确定协同环境-多基因 增加肌萎缩侧索硬化症风险的关联；2)定义肌萎缩侧索硬化表观基因组、转录组和 代谢组；以及3)确定环境暴露如何改变ALS免疫图谱和识别药物焦油- 得到了。首先，我们说明了来自自我报告、地理空间分析和生物群落的复杂暴露数据 使用特定暴露类型(例如农药、金属、空气污染)的组件-ERS(CER)和 用于联合曝光的ERS。我们使用多基因风险评分(PR)和C9ORF72来计算遗传风险 状态。我们将建立基于ERS和PRS的ALS风险和预测模型。接下来，使用CER、多ERS、 和PrS，我们测定了环境特征对DNA甲基组、mRNA和microRNA的影响，以鉴定 暴露与差异表达的基因和靶途径有关。表达数量性状 基因座(EQTL)分析将确定多态与暴露对基因表达的相对贡献。 高分辨率非靶向代谢组学将揭示ALS代谢组的环境特征和 识别新的毒物。所有数据集将使用泛组学技术进行整合，以确定基因代谢物 成为疾病目标的网络。最后，我们将对与CER和PolyR相关的免疫配置文件进行分类。 ERS使用FDA批准的现有药物确定治疗靶点。我们的建议具有很高的创新性；它使 对每个患者的罚款，(I)他们的暴露组，用CER/PolyERS汇总；(Ii)他们的基因组汇总如下 PR及其与ERS的关系，以了解基因/环境联合风险；(Iii)它们的多基因组学特征。 来自表观基因组、转录组、代谢组和炎症组的环境特征；(Iv)它们的失调- 根据它们与肌萎缩侧索硬化症风险和进展的关联进行排序，以确定个性化机制- 基于药物靶点；(V)通过危险因素修正的ALS预测模型和预防策略。我们的 并行、多组学方法比连续、单一毒物/组学方法快得多，而且它的集成度高。 磨砂自然捕捉到了组学交叉的全部光谱，加速了科学发现。我们会让 在寻找全新的治疗目标和公共卫生预防战略方面取得了重大进展。

ABSTRACT Genetic heritability incompletely explains amyotrophic lateral sclerosis (ALS), and the pace of ALS genetic dis- coveries has slowed, meaning entirely new research directions are needed to unravel disease mechanisms and identify therapies. Our goal is to understand, cure, and prevent ALS. Our overall approach is to identify the intersection of exposures, genomics, epigenomics, transcriptomics, metabolomics, and inflammation on ALS. Our rationale is that prior environmental risk scores (ERS) based on even crude plasma measures of limited classes of pollutants associate with a 7-fold increase in ALS risk and 2-fold decrease in survival, therefore a detailed understanding of the exposome with other omics can immediately provide new, much needed strate- gies for both ALS treatment and prevention. We propose 3 aims: 1) comprehensively assess environmental exposures and polygenic factors in ALS versus control subjects to identify synergistic environment-polygenic associations that increase ALS risk; 2) define exposome signatures in the ALS epigenome, transcriptome, and metabolome; and 3) determine how environmental exposures alter ALS immune profiles and identify drug tar- gets. First, we account for the complex exposure data from self-reports, geospatial analysis, and biospecimens using component-ERS (cERS) for specific exposure types (e.g. pesticides, metals, air pollution) and a poly- ERS for combined exposures. We account for genetic risk using polygenic risk scores (PRS) and C9ORF72 status. We will build ALS risk and prediction models based on ERS and PRS. Next, using cERS, poly-ERS, and PRS, we determine the environmental signature on the DNA methylome, mRNA and microRNA, to identify exposures that associate with differentially expressed genes and target pathways. Expression quantitative trait loci (eQTL) analyses will define the relative contribution of polymorphisms vs exposures on gene expression. High resolution untargeted metabolomics will reveal the environmental signature of the ALS metabolome and identify new toxicants. All datasets will be integrated using pan-omics techniques to identify gene-metabolite networks that are disease targets. Finally, we will classify immune profiles that associate with cERS and poly- ERS to identify therapeutic targets using existing FDA approved drugs. Our proposal is highly innovative; it de- fines for each patient, (i) their exposome, summarized with cERS/poly-ERS; (ii) their genome summarized by PRS and its association with ERS to understand the combined gene/environment risk; (iii) their multi-omic en- vironmental signatures from the epigenome, transcriptome, metabolome, and inflammasome; (iv) their dysreg- ulated pathways, ranked by their association to ALS risk and progression to identify personalized mechanism- based drug-targets; and (v) ALS prediction models and preventative strategies via risk factor modification. Our parallel, multi-omics approach is significantly faster than serial, single-toxicant/omic approaches, and its inte- grated nature captures the full spectrum of omics intersections, accelerating scientific discovery. We will make significant strides in finding completely new therapeutic targets and public health preventative strategies.