Physical Chemistry Seminar 228: Jennifer Dionne

Abstract: The oceans are the largest biological habitat in the known universe and are among the least charted. Covering over two and a half times the area of Mars, the oceans host diverse microorganisms that cycle nearly all chemical elements and are responsible for half of the global photosynthetic activity. Yet, studying the marine microbiome remains an outstanding challenge. Very few marine microbes have been successfully cultured under laboratory conditions, and culture-free methods like genomics and mass spectrometry are incompatible with the real time measurements necessary to study how physicochemical drivers impact microbial nutrient cycling. Here, we present our efforts to simultaneously and rapidly measure multiple ‘omic’ signatures from the ocean. First, we combine Raman spectroscopy and deep learning to accurately classify bacteria by both species and antibiotic resistance in a single step. With a convolutional neural network (CNN), we achieve species identification and antibiotic susceptibility accuracies similar to leading mass spectrometry techniques. Next, we describe resonant nanophotonic surfaces that enable detection of genes, proteins, and metabolites with femtomolar sensitivity. These metasurfaces produce a large amplification of the electromagnetic field intensity, increasing the response to minute refractive index changes from target binding; simultaneously, the light is beam-steered to particular detector pixels. By combining metasurface design with acoustic bioprinting for functionalization, we produce develop chips that detect gene fragments, proteins, and small molecule toxins on the same platform. We discuss integration of these sensors with autonomous underwater robots from Monterey Bay Aquarium Research Institute (MBARI) for real-time phytoplankton and phycotoxin detection.