On interaction of proteinoids with simulated neural networks

Abstract

Proteinoids are thermal proteins which swell into microspheres in solution. The proteinoid microspheres show spiking of electrical potential similar to that to that of living neurons. Rich spectrum of proteinoids spiking responses to optical and electrical stimulation makes them promising candidates for neuromorphic unconventional computing devices. We decided to evaluate neuron-like activity of proteinoids in the experimental setups of their interaction with simulate neuronal network of Izhikevich neurons. The simulated neural networks stimulate and modulate electrical activity of proteinoid ensembles by interacting with them via arbitrary form programmable function generator. Different amino-acid compositions of proteinoids responded uniquely to input spiking from simulated neurons. We demonstrated that patterns of electrical spiking activity of proteinoids and complexity of the activity can be tuned by patterns of spikes generated by simulated neurons. The research opens novel venues to establishing interacting between nanobrains - brain-like organoids made from molecules, not animal cells -- and real nervous systems.