Prediction of Metabolic Transformations using Cross Venn-ABERS Predictors

Staffan Arvidsson; Ola Spjuth; Lars Carlsson; Paolo Toccaceli

Prediction of Metabolic Transformations using Cross Venn-ABERS Predictors

Staffan Arvidsson, Ola Spjuth, Lars Carlsson, Paolo Toccaceli

Department of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Prediction of drug metabolism is an important topic in the drug discovery process, and we here present a study using probabilistic predictions applying Cross Venn-ABERS Predictors (CVAPs) on data for site-of-metabolism. We used a dataset of 73599 biotransformations, applied SMIRKS to define biotransformations of interest and constructed five datasets where chemical structures were represented using signatures descriptors. The results show that CVAP produces well-calibrated predictions for all datasets with good predictive capability, making CVAP an interesting method for further exploration in drug discovery applications.

Original language	English
Title of host publication	Proceedings of Machine Learning Research
Pages	118-131
Number of pages	14
Volume	60
Publication status	Published - 31 May 2017

Keywords

Venn-ABERS
Cross Venn-ABERS
Site-of-Metabolism
Drug discovery
Machine Learning
Support Vector Machine
Signatures descriptor
QSAR

Access to Document

http://proceedings.mlr.press/v60/arvidsson17a/arvidsson17a.pdf

Cite this

@inproceedings{5f2eb68dfcc043cb830a1a6fb079217d,

title = "Prediction of Metabolic Transformations using Cross Venn-ABERS Predictors",

abstract = "Prediction of drug metabolism is an important topic in the drug discovery process, and we here present a study using probabilistic predictions applying Cross Venn-ABERS Predictors (CVAPs) on data for site-of-metabolism. We used a dataset of 73599 biotransformations, applied SMIRKS to define biotransformations of interest and constructed five datasets where chemical structures were represented using signatures descriptors. The results show that CVAP produces well-calibrated predictions for all datasets with good predictive capability, making CVAP an interesting method for further exploration in drug discovery applications.",

keywords = "Venn-ABERS, Cross Venn-ABERS, Site-of-Metabolism, Drug discovery, Machine Learning, Support Vector Machine, Signatures descriptor, QSAR",

author = "Staffan Arvidsson and Ola Spjuth and Lars Carlsson and Paolo Toccaceli",

year = "2017",

month = may,

day = "31",

language = "English",

volume = "60",

pages = "118--131",

booktitle = "Proceedings of Machine Learning Research",

}

TY - GEN

T1 - Prediction of Metabolic Transformations using Cross Venn-ABERS Predictors

AU - Arvidsson, Staffan

AU - Spjuth, Ola

AU - Carlsson, Lars

AU - Toccaceli, Paolo

PY - 2017/5/31

Y1 - 2017/5/31

N2 - Prediction of drug metabolism is an important topic in the drug discovery process, and we here present a study using probabilistic predictions applying Cross Venn-ABERS Predictors (CVAPs) on data for site-of-metabolism. We used a dataset of 73599 biotransformations, applied SMIRKS to define biotransformations of interest and constructed five datasets where chemical structures were represented using signatures descriptors. The results show that CVAP produces well-calibrated predictions for all datasets with good predictive capability, making CVAP an interesting method for further exploration in drug discovery applications.

AB - Prediction of drug metabolism is an important topic in the drug discovery process, and we here present a study using probabilistic predictions applying Cross Venn-ABERS Predictors (CVAPs) on data for site-of-metabolism. We used a dataset of 73599 biotransformations, applied SMIRKS to define biotransformations of interest and constructed five datasets where chemical structures were represented using signatures descriptors. The results show that CVAP produces well-calibrated predictions for all datasets with good predictive capability, making CVAP an interesting method for further exploration in drug discovery applications.

KW - Venn-ABERS

KW - Cross Venn-ABERS

KW - Site-of-Metabolism

KW - Drug discovery

KW - Machine Learning

KW - Support Vector Machine

KW - Signatures descriptor

KW - QSAR

M3 - Conference contribution

VL - 60

SP - 118

EP - 131

BT - Proceedings of Machine Learning Research

ER -