Combining Machine Learning and Electrophysiology for Insect Odorant Receptor Studies

Insects rely on olfaction in many aspects of their life, and odorant receptors are key proteins in this process. Whereas a plethora of insect odorant receptor sequences is available, most of them are still orphan or uncompletely characterized, since their functional studies are usually limited by restricted odorant panels. With joint approaches that combine computational methods like machine learning and electrophysiology measurements, researchers can expand the chemical space of insect odorant receptors and speed up the discovery of new active ligands. This chapter details the methodology for setting up a quantitative structure-activity relationship (QSAR) predictive model for identifying odorant receptor agonists and for conducting single sensillum recordings to validate the predictions.