Predicting error bars for QSAR models

Timon Schroeter; Anton Schwaighofer; Sebastian Mika; Antonius Ter Laak; Detlev Suelzle; Ursula Ganzer; Nikolaus Heinrich; Klaus Robert Müller

doi:10.1063/1.2793398

Predicting error bars for QSAR models

Timon Schroeter, Anton Schwaighofer, Sebastian Mika, Antonius Ter Laak, Detlev Suelzle, Ursula Ganzer, Nikolaus Heinrich, Klaus Robert Müller

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

Abstract

Unfavorable physicochemical properties often cause drug failures. It is therefore important to take lipophilicity and water solubility into account early on in lead discovery. This study presents log D₇ models built using Gaussian Process regression, Support Vector Machines, decision trees and ridge regression algorithms based on 14556 drug discovery compounds of Bayer Schering Pharma. A blind test was conducted using 7013 new measurements from the last months. We also present independent evaluations using public data. Apart from accuracy, we discuss the quality of error bars that can be computed by Gaussian Process models, and ensemble and distance based techniques for the other modelling approaches.

Original language	English
Title of host publication	CompLife 2007 - 3rd International Symposium on Computational Life Science
Pages	158-167
Number of pages	10
DOIs	https://doi.org/10.1063/1.2793398
Publication status	Published - 2007
Event	3rd International Symposium on Computational Life Science, CompLife 2007 - Utrecht, Netherlands Duration: 2007 Oct 4 → 2007 Oct 5

Publication series

Name	AIP Conference Proceedings
Volume	940
ISSN (Print)	0094-243X
ISSN (Electronic)	1551-7616

Other

Other	3rd International Symposium on Computational Life Science, CompLife 2007
Country/Territory	Netherlands
City	Utrecht
Period	07/10/4 → 07/10/5

ASJC Scopus subject areas

General Physics and Astronomy

Access to Document

10.1063/1.2793398

Cite this

Schroeter, T, Schwaighofer, A, Mika, S, Ter Laak, A, Suelzle, D, Ganzer, U, Heinrich, N & Müller, KR 2007, Predicting error bars for QSAR models. in CompLife 2007 - 3rd International Symposium on Computational Life Science. AIP Conference Proceedings, vol. 940, pp. 158-167, 3rd International Symposium on Computational Life Science, CompLife 2007, Utrecht, Netherlands, 07/10/4. https://doi.org/10.1063/1.2793398

@inproceedings{04160773c355450c8a4602e4bfa15cd0,

title = "Predicting error bars for QSAR models",

abstract = "Unfavorable physicochemical properties often cause drug failures. It is therefore important to take lipophilicity and water solubility into account early on in lead discovery. This study presents log D7 models built using Gaussian Process regression, Support Vector Machines, decision trees and ridge regression algorithms based on 14556 drug discovery compounds of Bayer Schering Pharma. A blind test was conducted using 7013 new measurements from the last months. We also present independent evaluations using public data. Apart from accuracy, we discuss the quality of error bars that can be computed by Gaussian Process models, and ensemble and distance based techniques for the other modelling approaches.",

author = "Timon Schroeter and Anton Schwaighofer and Sebastian Mika and {Ter Laak}, Antonius and Detlev Suelzle and Ursula Ganzer and Nikolaus Heinrich and M{\"u}ller, {Klaus Robert}",

year = "2007",

doi = "10.1063/1.2793398",

language = "English",

isbn = "9780735404526",

series = "AIP Conference Proceedings",

pages = "158--167",

booktitle = "CompLife 2007 - 3rd International Symposium on Computational Life Science",

note = "3rd International Symposium on Computational Life Science, CompLife 2007 ; Conference date: 04-10-2007 Through 05-10-2007",

}

TY - GEN

T1 - Predicting error bars for QSAR models

AU - Schroeter, Timon

AU - Schwaighofer, Anton

AU - Mika, Sebastian

AU - Ter Laak, Antonius

AU - Suelzle, Detlev

AU - Ganzer, Ursula

AU - Heinrich, Nikolaus

AU - Müller, Klaus Robert

PY - 2007

Y1 - 2007

N2 - Unfavorable physicochemical properties often cause drug failures. It is therefore important to take lipophilicity and water solubility into account early on in lead discovery. This study presents log D7 models built using Gaussian Process regression, Support Vector Machines, decision trees and ridge regression algorithms based on 14556 drug discovery compounds of Bayer Schering Pharma. A blind test was conducted using 7013 new measurements from the last months. We also present independent evaluations using public data. Apart from accuracy, we discuss the quality of error bars that can be computed by Gaussian Process models, and ensemble and distance based techniques for the other modelling approaches.

AB - Unfavorable physicochemical properties often cause drug failures. It is therefore important to take lipophilicity and water solubility into account early on in lead discovery. This study presents log D7 models built using Gaussian Process regression, Support Vector Machines, decision trees and ridge regression algorithms based on 14556 drug discovery compounds of Bayer Schering Pharma. A blind test was conducted using 7013 new measurements from the last months. We also present independent evaluations using public data. Apart from accuracy, we discuss the quality of error bars that can be computed by Gaussian Process models, and ensemble and distance based techniques for the other modelling approaches.

UR - http://www.scopus.com/inward/record.url?scp=40249113486&partnerID=8YFLogxK

U2 - 10.1063/1.2793398

DO - 10.1063/1.2793398

M3 - Conference contribution

AN - SCOPUS:40249113486

SN - 9780735404526

T3 - AIP Conference Proceedings

SP - 158

EP - 167

BT - CompLife 2007 - 3rd International Symposium on Computational Life Science

T2 - 3rd International Symposium on Computational Life Science, CompLife 2007

Y2 - 4 October 2007 through 5 October 2007

ER -

Predicting error bars for QSAR models

Abstract

Publication series

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this