We have introduced Gene Regulatory Neural Networks (GRNN) as a novel interpretation of bacterial gene regulatory systems, mapping them into functional neural network models capable of computation within living cells. This innovative framework enables biological systems to perform machine learning tasks by leveraging the natural dynamics of gene expression and regulation.

Our GRNN architecture is designed to support wet-neuromorphic computing — a form of biocomputing where computation is carried out by living bacterial cells using gene-regulatory interactions. This approach not only offers adaptability and robustness but also drastically reduces energy consumption, making it highly suitable for low-power and edge AI applications.