OdoriFy: A conglomerate of artificial intelligence–driven prediction engines for olfactory decoding

Abstract

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Metabolic activity organizes olfactory representations

“…As shown in Figure 1b, POM does not show a significant or consistent advantage over generic chemical representations for predicting molecular properties that are not likely exploited by olfaction, such as electronic properties (e.g., QM7 39 ) and adverse drug reactions (e.g., SIDER 35 ) compiled by MoleculeNet 34 . We then apply POM to predict molecular binding activity for G-protein-coupled receptors (GPCRs, of which mammalian olfactory receptors are only a subset 29 ) generally, including those involved in enteric chemical sensation 40 (e.g., 5HT1A for serotonin and DRD2 for dopamine). While POM demonstrates superior performance for GPCRs involved in human olfaction, their performance is significantly worse for GPCRs related to enteric chemical sensation compared to generic chemical representations, showing specificity for olfaction (Figure 1b, Extended Data Figure 2).…”

“…This dataset is compiled from the literature and from databases as part of the OdoriFy effort 29 , and we use the binary response label for all eight different receptor targets including OR1A1, OR1A2, OR1G1, OR2J2, OR2W1, OR51E1, OR51E2, and OR52D1.…”

“…a) A graph neural network model pre-trained on human olfactory perceptual data produces a principal odor map, or POM (latent space, dashed box), which can be used to make predictions about any small, volatile molecule in biological and behavioral experiments. b) A random forest model using only POM produces predictions that meet or exceed those obtained from commonly-used generic molecular features32,33 (Mordred) across a range of olfactory datasets12,[22][23][24][25][26][27][28][29][30][31][34][35][36][37][38][39][40] in di erent species (green for ve ebrates and blue for inve ebrates), but not for prediction of non-odorous molecular prope ies (orange). The Y-axis is the di erence between pe ormance indices for models using POM vs. generic molecular features.…”

“…Here we pe orm a comprehensive meta analysis on 12 olfactory neuroscience datasets 12,[22][23][24][25][26][27][28][29][30][31] , spanning multiple species and levels of neural processing. We nd that the embedding from a graph neural network trained on human olfactory perception, which we term the principal odor map (POM), is highly predictive of the olfactory responses in nearly all datasets, even for species separated by hundreds of millions of years in evolution.…”

Artificial Intelligence uncovers carcinogenic human metabolites

“…Strikingly, an average validation involves harvesting of ~800 animals, raising ethical concerns 16,17 . While in vivo experiments offer indisputable identification of carcinogens, Artificial Intelligence can significantly accelerate pre-screening of the ever-expanding space of compounds that includes new drugs, chemicals, and industrial by-products [19][20][21][22] . To date, numerous computational methods have been proposed for carcinogenicity prediction.…”