POLYCYCLIC AROMATIC HYDROCARBONS: ULTRASENSITIVE DETECTION, EARLY LIFE EXPOSURES-CLINICAL OUTCOMES (PRETERM BIRTHS, CHRONIC LUNG DISEASE, AND NEUROCOGNITIVE DEFICITS), PREVENTION AND REMEDIATION

项目摘要 该项目的总体目标是开发新的和创新的方法， 检测和鉴定多环芳烃（PAH）分子及其官能化的 衍生物（多环芳香族化合物或PAC）。玛丽·巴约莫女士将从事这些领域的核心研究。 项目目标，由两个具体目标概述。本多样性补充的具体目标形成了一个中心， 实现项目目标所需工作的重要子集，并为 在本项目过程中开发的传感方法。它们也为将 机器学习方法进入光谱化学传感的子领域。在研究中， 化学和机器学习之间的接口为Ms. Bajomo，并将使她能够与三位PI及其研究小组在三个领域进行强有力的互动： 哈拉斯小组，在表面增强光谱传感领域的实验化学， Nordlander小组，基于纳米颗粒的基底设计，以及Patel小组，机器学习专家 数据科学我们的假设是，机器学习分类器可以被开发和使用， 区分环境或生物样品中发现的特定PAH和PAC分子， 它们的振动光谱特征这种方法的一个重要方面是识别PAH， PAC分子嵌入分子或聚合物捕获层， 来自环境和/或生物样品的PAH/PAC分子具有其自身的特定光谱特性， 签名“背景”。这些研究是检测多组分混合物的基础， PAH/PAC分子从现实的环境或生物样品，使用表面- 增强的光谱学和机器学习算法类似于杂乱， 复杂的背景环境。Bajomo女士将追求的两个具体目标是：具体目标1： 一种适用于多种PAH和PAC的通用捕获层的鉴定和定量表征 生物和环境样品中遇到的化合物。该捕获层将作为 用于表面增强拉曼和红外光谱基底的通用涂层， 从溶液中提取浓度适合检测的PAH和PAC分子，符合 这些化学物质的浓度在感兴趣的样品中发现。具体目标2：开发表面增强型 化学传感数据作为机器学习分类器的输入，以衡量它们在识别 利用ML方法对PAH/PAC分子进行了区分。这么近 实验光谱研究和ML分类器测试之间的协同作用提出了一个突出的 在两个非常重要和动态的研究领域之间的接口研究生培训的机会 给巴乔莫女士

Project Summary The overarching goal of this project is the development of new and innovative approaches to ultrasensitive detection and identification of polycyclic aromatic hydrocarbon (PAH) molecules and their functionalized derivatives (polycyclic aromatic compounds, or PACs). Ms. Mary Bajomo, will pursue research central to these project goals, as outlined by both Specific Aims. The Specific Aims of this Diversity Supplement form a central, essential subset of the work required to achieve the project goals, and provide an essential foundation for the sensing methods to be developed over the course of this project. They also lay the groundwork for bringing Machine Learning methods into the subfield of spectroscopic chemical sensing. Pursuing research at the interface between Chemistry and Machine Learning presents an exceptional training opportunity for Ms. Bajomo, and will allow her to interact strongly with the three PIs and their research groups in three fields: the Halas group, for experimental chemistry in the area of surface-enhanced spectroscopic sensing, the Nordlander group, on nanoparticle-based substrate design, and the Patel group, experts in Machine Learning and Data Science. Our hypothesis is that Machine Learning classifiers can be developed and used to distinguish between specific PAH and PAC molecules found in environmental or biological samples through their vibrational spectroscopic signatures. An essential aspect of this approach is the identification of PAH and PAC molecules while embedded in a molecular or polymer capture layer that has been designed to extract PAH/PAC molecules from environmental and/or biological samples that has its own specific spectroscopic signature “background”. These investigations are foundational to the detection of multicomponent mixtures of PAH/PAC molecules from realistic environmental or biological samples, using a combination of surface- enhanced spectroscopies and Machine Learning algorithms analogous to image recognition in cluttered, complex background environments. The two Specific Aims that Ms Bajomo will pursue are: Specific Aim 1: The identification and quantitative characterization of a universal capture layer for the wide range of PAH and PAC compounds encountered in biological and environmental samples. This capture layer would be serve as a universal coating for surface-enhanced Raman and Infrared spectroscopic substrates, and allow for the extraction of PAH and PAC molecules from solution in concentrations suitable for detection, consistent with concentrations of these chemicals found in samples of interest. Specific Aim 2: To develop surface-enhanced chemical sensing data as input to Machine Learning classifiers, to benchmark their effectiveness in identifying specific PAH molecules and distinguishing PAH/PAC molecules from each other by ML methods. This close synergy between experimental spectroscopic studies and ML classifier testing presents an outstanding opportunity for graduate training at the interface between two extremely important and dynamic research field for Ms. Bajomo.