Critical comparison of virtual screening methods against the MUV data set

Pekka Tiikkainen (Corresponding Author), Patrick Markt, Gerhard Wolber, Johannes Kirchmair, Simona Distinto, Antti Poso, Olli Kallioniemi

Research output: Contribution to journalArticleScientificpeer-review

34 Citations (Scopus)

Abstract

In the current work, we measure the performance of seven ligand-based virtual screening tools - five similarity search methods and two pharmacophore elucidators - against the MUV data set. For the similarity search tools, single active molecules as well as active compound sets clustered in terms of their chemical diversity were used as templates. Their score was calculated against all inactive and active compounds in their target class. Subsequently, the scores were used to calculate different performance metrics including enrichment factors and AUC values. We also studied the effect of data fusion on the results. To measure the performance of the pharmacophore tools, a set of active molecules was picked either random- or chemical diversity-based from each target class to build a pharmacophore model which was then used to screen the remaining compounds in the set. Our results indicate that template sets selected by their chemical diversity are the best choice for similarity search tools, whereas the optimal training sets for pharmacophore elucidators are based on random selection underscoring that pharmacophore modeling cannot be easily automated. We also suggest a number of improvements for future benchmark sets and discuss activity cliffs as a potential problem in ligand-based virtual screening.
Original languageEnglish
Pages (from-to)2168-2178
JournalJournal of Chemical Information and Modeling
Volume49
Issue number10
DOIs
Publication statusPublished - 2009
MoE publication typeA1 Journal article-refereed

Fingerprint

Screening
Ligands
performance
Molecules
Data fusion
Values

Cite this

Tiikkainen, P., Markt, P., Wolber, G., Kirchmair, J., Distinto, S., Poso, A., & Kallioniemi, O. (2009). Critical comparison of virtual screening methods against the MUV data set. Journal of Chemical Information and Modeling, 49(10), 2168-2178. https://doi.org/10.1021/ci900249b
Tiikkainen, Pekka ; Markt, Patrick ; Wolber, Gerhard ; Kirchmair, Johannes ; Distinto, Simona ; Poso, Antti ; Kallioniemi, Olli. / Critical comparison of virtual screening methods against the MUV data set. In: Journal of Chemical Information and Modeling. 2009 ; Vol. 49, No. 10. pp. 2168-2178.
@article{21edd3c0dd4b46f6a0977b3b9a781aab,
title = "Critical comparison of virtual screening methods against the MUV data set",
abstract = "In the current work, we measure the performance of seven ligand-based virtual screening tools - five similarity search methods and two pharmacophore elucidators - against the MUV data set. For the similarity search tools, single active molecules as well as active compound sets clustered in terms of their chemical diversity were used as templates. Their score was calculated against all inactive and active compounds in their target class. Subsequently, the scores were used to calculate different performance metrics including enrichment factors and AUC values. We also studied the effect of data fusion on the results. To measure the performance of the pharmacophore tools, a set of active molecules was picked either random- or chemical diversity-based from each target class to build a pharmacophore model which was then used to screen the remaining compounds in the set. Our results indicate that template sets selected by their chemical diversity are the best choice for similarity search tools, whereas the optimal training sets for pharmacophore elucidators are based on random selection underscoring that pharmacophore modeling cannot be easily automated. We also suggest a number of improvements for future benchmark sets and discuss activity cliffs as a potential problem in ligand-based virtual screening.",
author = "Pekka Tiikkainen and Patrick Markt and Gerhard Wolber and Johannes Kirchmair and Simona Distinto and Antti Poso and Olli Kallioniemi",
year = "2009",
doi = "10.1021/ci900249b",
language = "English",
volume = "49",
pages = "2168--2178",
journal = "Journal of Chemical Information and Modeling",
issn = "1549-9596",
publisher = "American Chemical Society ACS",
number = "10",

}

Tiikkainen, P, Markt, P, Wolber, G, Kirchmair, J, Distinto, S, Poso, A & Kallioniemi, O 2009, 'Critical comparison of virtual screening methods against the MUV data set', Journal of Chemical Information and Modeling, vol. 49, no. 10, pp. 2168-2178. https://doi.org/10.1021/ci900249b

Critical comparison of virtual screening methods against the MUV data set. / Tiikkainen, Pekka (Corresponding Author); Markt, Patrick; Wolber, Gerhard; Kirchmair, Johannes; Distinto, Simona; Poso, Antti; Kallioniemi, Olli.

In: Journal of Chemical Information and Modeling, Vol. 49, No. 10, 2009, p. 2168-2178.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Critical comparison of virtual screening methods against the MUV data set

AU - Tiikkainen, Pekka

AU - Markt, Patrick

AU - Wolber, Gerhard

AU - Kirchmair, Johannes

AU - Distinto, Simona

AU - Poso, Antti

AU - Kallioniemi, Olli

PY - 2009

Y1 - 2009

N2 - In the current work, we measure the performance of seven ligand-based virtual screening tools - five similarity search methods and two pharmacophore elucidators - against the MUV data set. For the similarity search tools, single active molecules as well as active compound sets clustered in terms of their chemical diversity were used as templates. Their score was calculated against all inactive and active compounds in their target class. Subsequently, the scores were used to calculate different performance metrics including enrichment factors and AUC values. We also studied the effect of data fusion on the results. To measure the performance of the pharmacophore tools, a set of active molecules was picked either random- or chemical diversity-based from each target class to build a pharmacophore model which was then used to screen the remaining compounds in the set. Our results indicate that template sets selected by their chemical diversity are the best choice for similarity search tools, whereas the optimal training sets for pharmacophore elucidators are based on random selection underscoring that pharmacophore modeling cannot be easily automated. We also suggest a number of improvements for future benchmark sets and discuss activity cliffs as a potential problem in ligand-based virtual screening.

AB - In the current work, we measure the performance of seven ligand-based virtual screening tools - five similarity search methods and two pharmacophore elucidators - against the MUV data set. For the similarity search tools, single active molecules as well as active compound sets clustered in terms of their chemical diversity were used as templates. Their score was calculated against all inactive and active compounds in their target class. Subsequently, the scores were used to calculate different performance metrics including enrichment factors and AUC values. We also studied the effect of data fusion on the results. To measure the performance of the pharmacophore tools, a set of active molecules was picked either random- or chemical diversity-based from each target class to build a pharmacophore model which was then used to screen the remaining compounds in the set. Our results indicate that template sets selected by their chemical diversity are the best choice for similarity search tools, whereas the optimal training sets for pharmacophore elucidators are based on random selection underscoring that pharmacophore modeling cannot be easily automated. We also suggest a number of improvements for future benchmark sets and discuss activity cliffs as a potential problem in ligand-based virtual screening.

U2 - 10.1021/ci900249b

DO - 10.1021/ci900249b

M3 - Article

VL - 49

SP - 2168

EP - 2178

JO - Journal of Chemical Information and Modeling

JF - Journal of Chemical Information and Modeling

SN - 1549-9596

IS - 10

ER -