Experimental and Computational Prediction of Glass Transition Temperature of Drugs
2014 (English)In: JOURNAL OF CHEMICAL INFORMATION AND MODELING, ISSN 1549-9596, Vol. 54, no 12, 3396-3403 p.Article in journal (Refereed) Published
Glass transition temperature (T-g) is an important inherent property of an amorphous solid material which is usually determined experimentally. In this study, the relation between T-g and melting temperature (T-m) was evaluated using a data set of 71 structurally diverse druglike compounds. Further, in silico models for prediction of T-g were developed based on calculated molecular descriptors and linear (multilinear regression, partial least-squares, principal component regression) and nonlinear (neural network, support vector regression) modeling techniques. The models based on T-m predicted T-g with an RMSE of 19.5 K for the test set. Among the five computational models developed herein the support vector regression gave the best result with RMSE of 18.7 K for the test set using only four chemical descriptors. Hence, two different models that predict T-g of drug-like molecules with high accuracy were developed. If T-m is available, a simple linear regression can be used to predict T-g. However, the results also suggest that support vector regression and calculated molecular descriptors can predict T-g with equal accuracy, already before compound synthesis.
Place, publisher, year, edition, pages
2014. Vol. 54, no 12, 3396-3403 p.
Computer Science Pharmacology and Toxicology
IdentifiersURN: urn:nbn:se:uu:diva-243056DOI: 10.1021/ci5004834ISI: 000347137500011PubMedID: 25361075OAI: oai:DiVA.org:uu-243056DiVA: diva2:787788