logS: The aqueous solubility data

In wind22zhu/enpls1.2: Ensemble Partial Least Squares Regression

Description

The aqueous solubility data reported by Hou et al.

Usage

data(logS)

Format

A list with 6 components:

• x - data frame with 549 rows (samples) and 166 columns (predictors)

• y - numeric vector of length 549 (response)

• x.test1 - data frame with 274 rows (samples) and 166 columns (predictors)

• y.test1 - numeric vector of length 274 (response)

• x.test2 - data frame with 446 rows (samples) and 166 columns (predictors)

• y.test2 - numeric vector of length 446 (response)

Details

This dataset contains 549 aqueous solubility samples (y) which have been modeled by 166 molecular descriptors (x) for training set, 274 aqueous solubility samples (y) which have been modeled by 166 molecular descriptors (x) for test1 set and 446 aqueous solubility samples (y) which have been modeled by 166 molecular descriptors (x) for test2 set.

Molecular descriptor types:

• Constitution

• Topology

• Connectivity

• Kappa

• EState

• Autocorrelation-moran

• Autocorrelation-geary

• Autocorrelation-broto

• Molecular properties

• Charge

• Moe-Type descriptors

References

Hou, T. J.; Xia, K.; Zhang, W.; Xu, X. J. "ADME Evaluation in Drug Discovery. 4. Prediction of Aqueous Solubility Based on Atom Contribution Approach" Journal of chemical information and computer sciences 44, no. 1 (2004): 266-275.

Examples

data(logS)
str(logS)

wind22zhu/enpls1.2 documentation built on May 4, 2019, 6:31 a.m.