The taste of a chemical compound present in food stimulates us to take in nutrients and avoid poisons. Many active ingredients present in drugs taste bitter and thus are aversive to children as well as many adults. Bitterness of medicines presents compliance problems and early flagging of potential bitterness of a drug candidate may help its further development. Taste prediction of a compound is of large interest for the food industry.
In this work, we have built 24 different machine learning models (Using 8 sampling methods) to predict three different taste endpoints - bitterness, sweetness, and sourness of chemical compounds. To predict a compound, please click the prediction button. For information on how to use the webserver and interpret the results, please refer to the FAQ.
We have analysed and validated our models using eight different sampling methods and four different machine learning algorithms.Two different feature based fingerprints are used to build respective models. To know more about our models, method and data set used please go to model information page.
Bitter tasting drugs are a major concern of compliance for children. Sensory tasting of drug candidates by humans is not a trivial matter, since it requires ethical approval, which is achievable only after a thorough toxicological study. Thus, efficient prediction of compounds sweetness as well as bitterness is not only of great interest to the nutrition industry and basic taste research but also for the drug discovery process. We have computed the taste predictions of approved drugs using our models. Please check Drug Taste section or click view details.