Machine Learning-Guided Decoding Bioelectronic Signals of Photosynthetic Cyanobacterial Cells by Conducting Polymers

Rapid and reliable communication with photosynthetic organisms is essential for monitoring their physiological states, which would reflect their endogenous bioelectric signals. However, these signals are inherently weak and challenging to detect due to the low efficiency of bioelectronic generation and transmission. In this work, we constructed a Syne/PFP/PPy biohybrid system by integrating the photosynthetic cyanobacterial cells of Synechococcus sp. PCC7942 (Syne) with cationic poly(fluorene-co-phenylene) derivative (PFP) and polypyrrole (PPy). Leveraging the superior light-harvesting ability of PFP and the excellent electrical conducting properties of PPy, the Syne/PFP/PPy system can enhance the acquisition of in situ bioelectronic signals from Syne. Compared to single Syne cells, the intensity of bioelectronic signals collected by Syne/PFP/PPy increased over 14 times (from 3.4 to 47.9 nA cm^-2). By incorporating machine learning models, we successfully correlated these bioelectronic signals with key physiological conditions, including variations in temperature, light intensity, pH, and nutrient availability (nitrogen and phosphorus). Furthermore, a wireless device was also designed for a simulation application of this system to realize wireless monitoring of Syne cells. This strategy offers a promising platform for decoding intrinsic bioelectronic signals of photosynthetic organisms and providing an efficient method for timely tracking their life status.