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

点击翻译按钮获取中文摘要

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.