Source: UCI Machine Learning Repository
Content
A dataset of manually-curated BCF for 779 chemicals was used to determine the mechanisms of bioconcentration, i.e. to predict whether a chemical: (1) is mainly stored within lipid tissues, (2) has additional storage sites (e.g. proteins), or (3) is metabolized/eliminated. Data were randomly split into a training set of 584 compounds (75) and a test set of 195 compounds (25), preserving the proportion between the classes. Two QSAR classification trees were developed using CART (Classification and Regression Trees) machine learning technique coupled with Genetic Algorithms. The file contains the selected Dragon descriptors (9) along with CAS, SMILES, experimental BCF, experimental/predicted KOW and mechanistic class (1, 2, 3). Further details on model development and performance along with descriptor definitions and interpretation are provided in the original manuscript (Grisoni et al., 2016).
Relevant Papers:
F. Grisoni, V.Consonni, M.Vighi, S.Villa, R.Todeschini (2016). Investigating the mechanisms of bioconcentration through QSAR classification trees, Environment International, 88, 198-205
Citation Request:
The dataset is freeware and may be used if proper reference is given to the authors. Please, refer to the following papers:
F. Grisoni, V.Consonni, M.Vighi, S.Villa, R.Todeschini (2016). Investigating the mechanisms of bioconcentration through QSAR classification trees, Environment International, 88, 198-205.
F. Grisoni, V. Consonni, S. Villa, M. Vighi, R. Todeschini (2015). QSAR models for bioconcentration: Is the increase in the complexity justified by more accurate predictions?. Chemosphere, 127, 171-179.