|
|
|
|
|
|
| About Yojiro Sakiyama (Pfizer) |
|
Dr. Yojiro Sakiyama received his Ph.D. in Biomedical Science from the University of Tokyo in 1993. During his post-doc he pursued non-invasive in vivo measurement using animal PET in the Tokyo Metropolitan Institute of Gerontology and National Institute of Longevity Sciences. During this time he investigated various methodologies including quantitative analysis of the kinetics of radioactive compounds. He joined Pfizer Nagoya Laboratories in 1999. During the first five years he investigated various automated measurement techniques as well as various novel statistical analysis methods for the efficient process of in vivo animal studies. Since 2004 his support extended to medicinal chemists, to apply recent data mining tools to their data to optimize the process of drug discovery. Since 2007, he joined Pfizer Sandwich Laboratories in the United Kingdom where his main research field is now in silico ADME modelling. His main current interest is especially to focus on the application of recent new machine learning tools for prediction of ADME endpoints.
|
|
Application of Machine Learning Tools for in silico ADME Screening
Yojiro Sakiyama, Pfizer Global Research and Development, Pfizer Inc., IPC654 Ramsgate Road, Sandwich, Kent CT13 9NJ, UK
To deal with the vast quantity of data from large compound libraries, computational in silico ADME screening is required to maximize efficiency of the drug discovery process. On the other hand, various machine learning tools originating from engineering areas have gradually gained considerable interest in drug discovery research. Here we have derived a relationship between the chemical structure and its ADME properties for a data set of in-house compounds by means of various in silico machine learning tools such as random forest, support vector machine, logistic regression and recursive partitioning. For model building, proprietary compounds comprising two classes (stable/unstable) were used with molecular descriptors calculated by the Molecular Operating Environment. The results using test compounds have demonstrated satisfactory results. Above all, classification by random forest as well as support vector machine yielded satisfactory results in an independent validation set, suggesting that nonlinear/ensemble-based classification methods might prove useful in the area of in silico ADME modeling.
|
|
|
|
|
|
|
|
|
