Elucidating causes of vaginal symptoms using a multi-omics approach

 描述(申请人提供)：阴道分泌物、灼热、瘙痒和异味是育龄妇女报告的最常见的阴道症状，仅在美国每年就有数百万人因此就诊。这些症状会造成严重的不适，经常反复出现，并对女性的自尊和生活质量产生负面影响。通常，女性会诉诸不明智的替代治疗，试图改善她们的症状，这些症状只会提供短暂的缓解，在许多情况下会增加症状的严重性。这些症状与许多严重的妇产科后果有关，包括增加性传播感染的风险。这项拟议工作的长期目标是更全面地了解阴道症状的根本原因，以便制定有针对性的有效策略，有效地预防或治疗阴道分泌物、不适和异味。我们建议收集和分析的多重组学数据集将使我们能够检验中心假设，即症状是由特定功能相互作用产生的阴道微生物群的紧急特性，而不仅仅是引起这些体征和症状的阴道微生物群的微生物组成和宿主反应。为了实现这一目标，我们将利用一组独特的阴道拭子样本和广泛的元数据，这些样本由135名女性在10周内每天前瞻性地收集。在这项研究中，女性经历了1)明显的阴道症状，2)慢性和持续性症状，或3)没有症状。这项研究设计提供了一个独特的机会，可以对在阴道症状发作之前、期间和之后收集的样本使用组学技术，并将这些样本与有慢性症状或没有症状的妇女进行比较，并确定特定的预测生物标志物，这些生物标志物将转化为对妇女健康的更个性化管理。拟议的项目将针对两个具体目标：(1)通过表征在阴道症状发作之前、期间和之后收集的阴道样本中的细菌元基因组、宿主和细菌转录体和代谢体以及天然免疫标记物来确定阴道微生物组的组成和功能，并将这些数据与无症状妇女的可比数据进行比较。(2)综合运用系统生物学方法，运用套索回归、岭回归和拉格朗日网络回归等多管齐下的建模策略，结合稳健的现代模型选择技术和网络分析，建立起病的预测和因果模型，实现稳健性、解释力和规模(预测变量的数量)之间的平衡。有了对阴道症状发生和恢复过程中微生物区系和宿主基因组因子变化的详细了解，将有可能开发出更准确和敏感的诊断程序、新的治疗策略和确保阴道动态平衡的有效手段。

 DESCRIPTION (provided by applicant): Vaginal discharge, burning, itching, and malodor are the most common vaginal symptoms reported by reproductive-age women resulting in millions of health care visits annually in the United States alone. These symptoms create substantial discomfort, are often recurrent, and negatively impact women's self- esteem and quality of life. Often, women resort to ill-advised alternative treatments in an attempt to ameliorate their symptoms that only provide a short-lived relief, and in many cases increase symptom severity. These symptoms are associated with numerous serious gynecological and obstetric outcomes, including an increased risk for sexually transmitted infections. The long-term goal of the proposed work is to develop a more complete understanding of the underlying causes of vaginal symptoms so that targeted and effective strategies can be developed to effectively prevent or treat vaginal discharge, discomfort, and malodor. The multi-omics datasets we propose to gather and analyze will allow us to test the central hypothesis that symptoms are emergent properties of the vaginal microbiome that result from the interplay of specific functions, and not simply microbial composition of vaginal microbiomes and host responses elicit these signs and symptoms. To achieve this, we will leverage a unique set of vaginal swabs samples and extensive metadata that were prospectively collected daily by 135 women for 10 weeks. During this study, women experienced either 1) clear episodes of vaginal symptoms, 2) chronic and persistent symptoms or 3) no symptoms. The study design affords a unique opportunity to use `omics technologies on samples collected before, during and after episode of vaginal symptoms and compare these to women with chronic or no symptoms, and identify specific predictive biomarkers that will translate to more personalized management of women's health. The proposed project will address two specific aims: (1) Determine the composition and function of the vaginal microbiome by characterizing bacterial metagenomes, the host and bacterial transcriptomes and metabolomes, and markers of innate immunity in vaginal samples collected prior to, during and after episodes of vaginal symptoms and compare these to comparable data from asymptomatic women. (2) Develop predictive and causal models of symptom onset using integrative systems biology approaches using a multi- prong modeling strategy that includes lasso regression, ridge regression and elactic net regression combined with robust and modern model selection techniques, and network analyses, we will achieve a predictive and causal model that is a balance of robustness, explanatory power, and size (in terms of the number of predictor variables). Armed with detailed knowledge of changes in genomic factor of the microbiota and the host during the onset and recovery from vaginal symptoms it will be possible to develop more accurate and sensitive diagnostic procedures, new therapeutic strategies, and effective means to insure vaginal homeostasis.