Using Conformal Prediction to Prioritize Compound Synthesis in Drug Discovery

Ernst Ahlberg, Susanne Winiwarter, Henrik Boström, Henrik Linusson, Tuve Löfström, Ulf Norinder Ulf Johansson, Ola Engkvist, Oscar Hammar, Claus Bendtsen, Lars Carlsson
Proceedings of the Sixth Workshop on Conformal and Probabilistic Prediction and Applications, PMLR 60:174-184, 2017.

Abstract

The choice of how much money and resources to spend to understand certain problems is of high interest in many areas. This work illustrates how computational models can be more tightly coupled with experiments to generate decision data at lower cost without reducing the quality of the decision. Several different strategies are explored to illustrate the trade off between lowering costs and quality in decisions. AUC is used as a performance metric and the number of objects that can be learnt from is constrained. Some of the strategies described reach AUC values over 0.9 and outperforms strategies that are more random. The strategies that use conformal predictor p-values show varying results, although some are top performing.

Cite this Paper

BibTeX
@InProceedings{pmlr-v60-ahlberg17a, title = {Using Conformal Prediction to Prioritize Compound Synthesis in Drug Discovery}, author = {Ahlberg, Ernst and Winiwarter, Susanne and Boström, Henrik and Linusson, Henrik and Löfström, Tuve and Ulf Johansson, Ulf Norinder and Engkvist, Ola and Hammar, Oscar and Bendtsen, Claus and Carlsson, Lars}, booktitle = {Proceedings of the Sixth Workshop on Conformal and Probabilistic Prediction and Applications}, pages = {174--184}, year = {2017}, editor = {Gammerman, Alex and Vovk, Vladimir and Luo, Zhiyuan and Papadopoulos, Harris}, volume = {60}, series = {Proceedings of Machine Learning Research}, month = {13--16 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v60/ahlberg17a/ahlberg17a.pdf}, url = {https://proceedings.mlr.press/v60/ahlberg17a.html}, abstract = {The choice of how much money and resources to spend to understand certain problems is of high interest in many areas. This work illustrates how computational models can be more tightly coupled with experiments to generate decision data at lower cost without reducing the quality of the decision. Several different strategies are explored to illustrate the trade off between lowering costs and quality in decisions. AUC is used as a performance metric and the number of objects that can be learnt from is constrained. Some of the strategies described reach AUC values over 0.9 and outperforms strategies that are more random. The strategies that use conformal predictor p-values show varying results, although some are top performing.} }
Endnote
%0 Conference Paper %T Using Conformal Prediction to Prioritize Compound Synthesis in Drug Discovery %A Ernst Ahlberg %A Susanne Winiwarter %A Henrik Boström %A Henrik Linusson %A Tuve Löfström %A Ulf Norinder Ulf Johansson %A Ola Engkvist %A Oscar Hammar %A Claus Bendtsen %A Lars Carlsson %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-ahlberg17a %I PMLR %P 174--184 %U https://proceedings.mlr.press/v60/ahlberg17a.html %V 60 %X The choice of how much money and resources to spend to understand certain problems is of high interest in many areas. This work illustrates how computational models can be more tightly coupled with experiments to generate decision data at lower cost without reducing the quality of the decision. Several different strategies are explored to illustrate the trade off between lowering costs and quality in decisions. AUC is used as a performance metric and the number of objects that can be learnt from is constrained. Some of the strategies described reach AUC values over 0.9 and outperforms strategies that are more random. The strategies that use conformal predictor p-values show varying results, although some are top performing.
APA
Ahlberg, E., Winiwarter, S., Boström, H., Linusson, H., Löfström, T., Ulf Johansson, U.N., Engkvist, O., Hammar, O., Bendtsen, C. & Carlsson, L.. (2017). Using Conformal Prediction to Prioritize Compound Synthesis in Drug Discovery. Proceedings of the Sixth Workshop on Conformal and Probabilistic Prediction and Applications, in Proceedings of Machine Learning Research 60:174-184 Available from https://proceedings.mlr.press/v60/ahlberg17a.html.

Related Material