A status report on human odorant receptors and their allocated agonists.

Olfactory perception begins when odorous substances interact with specialized receptors located on the surface of dedicated sensory neurons. The recognition of smells depends on a complex mechanism involving a combination of interactions between an odorant and a set of odorant receptors (ORs), where molecules are recognized according to a combinatorial activation code of ORs. Although these interactions have been studied for decades, the rules governing this ligand recognition remain poorly understood, and the complete combinatorial code is only known for a handful of odorants. We have carefully analyzed experimental results regarding the interactions between ORs and molecules to provide a status report on the deorphanization of ORs, i.e. the identification of the first agonist for a given sequence. This meticulous analysis highlights the influence of experimental methodology (cell line or readout) on molecule-receptor association results and shows that 83% of the results are conserved regardless of experimental conditions. The distribution of another key parameter, EC50, indicates that most OR ligand activities are in the micromolar range and that impurities could lead to erroneous conclusions. Focusing on the human ORs, our study shows that 88% of the documented sequences still need to be deorphanized. Finally, we also estimate the size of the ORs' recognition range, or broadness, as the number of odorants activating a given OR. By analogously estimating molecular broadness and combining the two estimates we propose a basic framework that can serve as a comparison point for future machine learning algorithms predicting OR-molecule activity.