Prediction of Metabolic Transformations using Cross Venn-ABERS Predictors

Staffan Arvidsson, Ola Spjuth, Lars Carlsson, Paolo Toccaceli
; Proceedings of the Sixth Workshop on Conformal and Probabilistic Prediction and Applications, PMLR 60:118-131, 2017.

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.

Cite this Paper


BibTeX
@InProceedings{pmlr-v60-arvidsson17a, title = {Prediction of Metabolic Transformations using Cross {V}enn-{ABERS} Predictors}, author = {Staffan Arvidsson and Ola Spjuth and Lars Carlsson and Paolo Toccaceli}, booktitle = {Proceedings of the Sixth Workshop on Conformal and Probabilistic Prediction and Applications}, pages = {118--131}, year = {2017}, editor = {Alex Gammerman and Vladimir Vovk and Zhiyuan Luo and Harris Papadopoulos}, volume = {60}, series = {Proceedings of Machine Learning Research}, address = {Stockholm, Sweden}, month = {13--16 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v60/arvidsson17a/arvidsson17a.pdf}, url = {http://proceedings.mlr.press/v60/arvidsson17a.html}, 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.} }
Endnote
%0 Conference Paper %T Prediction of Metabolic Transformations using Cross Venn-ABERS Predictors %A Staffan Arvidsson %A Ola Spjuth %A Lars Carlsson %A Paolo Toccaceli %B Proceedings of the Sixth Workshop on Conformal and Probabilistic Prediction and Applications %C Proceedings of Machine Learning Research %D 2017 %E Alex Gammerman %E Vladimir Vovk %E Zhiyuan Luo %E Harris Papadopoulos %F pmlr-v60-arvidsson17a %I PMLR %J Proceedings of Machine Learning Research %P 118--131 %U http://proceedings.mlr.press %V 60 %W PMLR %X 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.
APA
Arvidsson, S., Spjuth, O., Carlsson, L. & Toccaceli, P.. (2017). Prediction of Metabolic Transformations using Cross Venn-ABERS Predictors. Proceedings of the Sixth Workshop on Conformal and Probabilistic Prediction and Applications, in PMLR 60:118-131

Related Material