Predicting lipophilicity of drug-discovery molecules using Gaussian process models

Schroeter, T.S.T.S.SchroeterSchwaighofer, A.A.SchwaighoferMika, S.S.MikaTer Laak, A.A.Ter LaakSülzle, D.D.SülzleGanzer, U.U.GanzerHeinrich, N.N.HeinrichMüller, K.-R.K.-R.Müller2022-03-042022-03-042007https://publica.fraunhofer.de/handle/publica/21515710.1002/cmdc.200700041The lipophilicity of 14556 library compounds at Bayer Schering was modeled using Gaussian process methodology. In a blind test with 7013 new drug-discovery molecules from the last few months, 81% were predicted correctly within one log unit,compared with only 44% achieved by commercial software. Predicted error bars exhibit close to ideal statistical properties, thereby allowing assessment of the model's domain of applicability.en004615Predicting lipophilicity of drug-discovery molecules using Gaussian process modelsjournal article