Optibrium Ltd., a provider of software and AI solutions for drug discovery, announced the publication of “Imputation of Sensory Properties Using Deep Learning”* in the Journal of Computer-Aided Molecular Design.
The peer-reviewed study, conducted in collaboration by International Flavors and Fragrances (IFF) and Intellegens, demonstrated high predictive reliability for Optibrium’s augmented chemistry on human panel-based assessments, such as odor intensity and odor detection threshold. This presents opportunities to reduce the need for human testing when developing new flavor and fragrance ingredients.
Related: Cross-cultural Influences in Multi-sensory Flavor Perception and Association
Ingredient development for fragrances currently requires human panel-based trials, according to the study. Augmented chemistry draws on data for the property of interest and utilizes measurements and data from other endpoints to make predictions. This increases the predictive power for the odor perception threshold by learning the relationships between different experiments and using earlier-stage measurements to make more accurate predictions.
The study demonstrated translational capabilities that expand the potential computational approaches to late-stage discovery methods that are resource-intensive, like human trial data. It offers opportunities to reduce the time and cost of developing new flavors and fragrances. It also offers new insights into drug discovery data.
Samar Mahmoud, senior scientist at Optibrium, said, “We have demonstrated augmented chemistry’s unprecedented capabilities to robustly predict otherwise intractable complex endpoints in several drug discovery collaborations with leading pharma and biotech companies. This study further underlines its translational capabilities, providing further evidence for its applicability to other related trial outcomes across fields such as drug discovery.”
Dmitriy Chekmarev, senior research investigator at IFF, said, “We were impressed by the outcomes of the collaboration. Not just because it is the first approach that offers meaningful overall predictive power, but because its reliable uncertainty estimates give us the confidence to make critical project decisions based on computational methods.”
*Deep imputation on large-scale drug discovery data