Cellular function is orchestrated by intricate molecular interactions, and recently, the study of membrane-less organelles has unveiled a new layer of complexity in cell biology. Liquid-liquid phase separation (LLPS) plays a pivotal role in cellular processes such as signal transduction, gene regulation, and synaptic function. This project aims to harness the power of artificial intelligence (AI) and neuronal networks to predict and elucidate the LLPS behavior of synaptic proteins forming distinct membrane-less organelles.

We will employ AI algorithms and neuronal networks to build predictive models capable of identifying protein partners with LLPS potential. This will involve both supervised and unsupervised learning approaches. Extracting relevant features from the protein sequences, including amino acid composition, charge distribution, hydrophobicity patterns, and known LLPS-driving motifs will allow to identify new LLPS systems. We will then interpret the predictions made by the AI models to identify and test motifs and biophysical processes associated with synaptic LLPS behavior.

This project holds the potential to revolutionize our understanding of synaptic biology and protein phase separation. The insights gained from this study have several broad-reaching implications. It will unveil new mechanisms underlying protein phase separation at the synapse and it will allow to understand new relationships between protein phase separation and neurological disorders. This project will contribute to the advancement of scientific knowledge by addressing a significant gap in our understanding of synaptic protein organization and its role in synaptic health.