Deonarine Andrew, Batwara Ayushi, Wada Roy, Sharma Puneet, Loscalzo Joseph, Ojikutu Bisola, Hall Kathryn
Boston Public Health Commission, 1010 Massachusetts Avenue, 6th Floor, Boston, MA 02118, USA; School of Population and Public Health, University of British Columbia, 2206 East Mall, Vancouver, BC V6T 1Z3, Canada; Icahn School of Medicine at Mount Sinai, 1 Gustave L. Levy Pl, New York, NY 10029, USA.
Boston Public Health Commission, 1010 Massachusetts Avenue, 6th Floor, Boston, MA 02118, USA; University of California, Berkeley, 110 Sproul Hall #5800, Berkeley, CA 94720-5800, USA.
EBioMedicine. 2025 Feb;112:105575. doi: 10.1016/j.ebiom.2025.105575. Epub 2025 Jan 31.
Determining spatial relationships between diseases and the exposome is limited by available methodologies. aPEER (algorithm for Projection of Exposome and Epidemiological Relationships) uses machine learning (ML) and network analysis to find spatial relationships between diseases and the exposome in the United States.
Using aPEER we examined the relationship between 12 chronic diseases and 186 pollutants. PCA, K-means clustering, and map projection produced clusters of counties derived from pollutants, and the Jaccard correlation between these clusters with chronic disease geography (defined as groups of counties with high chronic disease prevalence rates) was calculated. Disease-pollution correlation matrices were used together with network analysis to identify the strongest disease-pollution relationships. Results were compared to LISA, Moran's I, univariate, elastic net, and random forest regression.
aPEER produced 68,820 human interpretable maps with distinct pollution-derived regions, and acetaldehyde/benzo(a)pyrene was found to be strongly associated with hypertension (J = 0.5316, p = 3.89 × 10), stroke (J = 0.4517, p = 1.15 × 10), and diabetes mellitus (J = 0.4425, p = 2.34 × 10); formaldehyde/glycol ethers with COPD (J = 0.4545, p = 8.27 × 10); and acetaldehyde/formaldehyde with stroke mortality (J = 0.4445, p = 4.28 × 10). Methanol, acetaldehyde, and formaldehyde formed distinct regions in the southeast United States (which correlated with both the Stroke and Diabetes Belts) which were strongly associated with multiple chronic diseases. Pollutants predicted chronic disease geography with similar or superior areas under the curve compared to SDOH and preventive healthcare models (determined with random forest and elastic net methods). Conventional geospatial analysis methods did not identify these geospatial relationships, highlighting aPEER's utility.
aPEER identified a pollution-defined geographical region associated with chronic disease, highlighting the role of aPEER in epidemiological and geospatial analysis, and exposomics in understanding chronic disease geography.
This work was primarily funded by the BPHC, NHLBI (R03 HL157890) and the CDC, and this work was funded in part by grants from the NIH (U01 HG007691, R01 HL155107, and HL166137), the American Heart Association (AHA24MERIT1185447), and the EU (HorizonHealth 2021 101057619) to JL.
现有方法限制了对疾病与暴露组之间空间关系的确定。aPEER（暴露组与流行病学关系投影算法）利用机器学习（ML）和网络分析来寻找美国疾病与暴露组之间的空间关系。
使用aPEER，我们研究了12种慢性病与186种污染物之间的关系。主成分分析（PCA）、K均值聚类和地图投影生成了由污染物衍生的县集群，并计算了这些集群与慢性病地理分布（定义为慢性病患病率高的县组）之间的杰卡德相关性。疾病-污染相关矩阵与网络分析一起用于识别最强的疾病-污染关系。将结果与局部空间自相关（LISA）、莫兰指数I、单变量、弹性网络和随机森林回归进行比较。
aPEER生成了68820张具有不同污染衍生区域的可人工解释地图，发现乙醛/苯并(a)芘与高血压（J = 0.5316，p = 3.89×10）、中风（J = 0.4517，p = 1.15×10）和糖尿病（J = 0.4425，p = 2.34×10）密切相关；甲醛/乙二醇醚与慢性阻塞性肺疾病（COPD）（J = 0.4545，p = 8.27×10）密切相关；乙醛/甲醛与中风死亡率（J = 0.4445，p = 4.28×10）密切相关。甲醇, 乙醛和甲醛在美国东南部形成了不同的区域（与中风带和糖尿病带相关），这些区域与多种慢性病密切相关。与社会人口统计学和健康相关因素（SDOH）及预防性医疗保健模型相比，污染物预测慢性病地理分布的曲线下面积相似或更大（通过随机森林和弹性网络方法确定）。传统的地理空间分析方法未识别出这些地理空间关系，凸显了aPEER的实用性。
aPEER识别出了一个与慢性病相关的污染定义地理区域，凸显了aPEER在流行病学和地理空间分析以及暴露组学在理解慢性病地理分布方面的作用。
这项工作主要由BPHC、NHLBI（R03 HL157890）和疾病预防控制中心资助，这项工作部分由美国国立卫生研究院（U01 HG007691、R01 HL155107和HL166137）、美国心脏协会（AHA24MERIT1185447）和欧盟（HorizonHealth 2021 101057619）授予JL的赠款资助。
