Photonic Data Science

Quantitative measurement is essential for modern bioreactor synthesis, which plays a crucial role in the production of foods and pharmaceuticals. These complex processes rely on living organisms, and require fast, high-performing inline analytics for effective process control and intervention. In this study, a glycerol-fed, lab-scale E. coli bioprocess producing representative pharmaceutical compounds was monitored offline with a portable, high-sensitivity Raman spectrometer. Sampling and high performance liquid chromatography (HPLC) analysis provided the ground truth calibration for a chemometric model developed using an easy-to-use, built-for-Raman chemometric software package. The chemometric model developed demonstrates the ability of this approach to offer spectroscopic prediction of feedstock, active pharmaceutical ingredients (API), and side product concentrations. This specific combination of compact spectrometer and AI-driven analysis software opens the door for the application of real-time, continuous, and non-invasive Raman spectroscopy in bioprocess monitoring, even for users without prior chemometric or Raman experience.