Limits...
A ranking method for the concurrent learning of compounds with various activity profiles.

Dörr A, Rosenbaum L, Zell A - J Cheminform (2015)

Bottom Line: The experiments show that ranking-based algorithms show an increased performance for single- and multi-target virtual screening.Moreover, compounds that do not completely fulfill the desired activity profile are still ranked higher than decoys or compounds with an entirely undesired profile, compared to other multi-target SVM methods.SVM-based ranking methods constitute a valuable approach for virtual screening in multi-target drug design.

View Article: PubMed Central - PubMed

Affiliation: Center for Bioinformatics Tübingen (ZBIT), University of Tuebingen, Sand 1, Tübingen, 72076 Germany.

ABSTRACT

Background: In this study, we present a SVM-based ranking algorithm for the concurrent learning of compounds with different activity profiles and their varying prioritization. To this end, a specific labeling of each compound was elaborated in order to infer virtual screening models against multiple targets. We compared the method with several state-of-the-art SVM classification techniques that are capable of inferring multi-target screening models on three chemical data sets (cytochrome P450s, dehydrogenases, and a trypsin-like protease data set) containing three different biological targets each.

Results: The experiments show that ranking-based algorithms show an increased performance for single- and multi-target virtual screening. Moreover, compounds that do not completely fulfill the desired activity profile are still ranked higher than decoys or compounds with an entirely undesired profile, compared to other multi-target SVM methods.

Conclusions: SVM-based ranking methods constitute a valuable approach for virtual screening in multi-target drug design. The utilization of such methods is most helpful when dealing with compounds with various activity profiles and the finding of many ligands with an already perfectly matching activity profile is not to be expected.

No MeSH data available.


Related in: MedlinePlus

Performance of the cytochrome P450s single-target data set with binary test sets. Each boxplot depicts the mean ranking error on the 20 randomly generated test sets for each target separated according to the secondary target that should be avoided with higher priority. The given ranking error is equal to 1−AUC.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4306736&req=5

Fig10: Performance of the cytochrome P450s single-target data set with binary test sets. Each boxplot depicts the mean ranking error on the 20 randomly generated test sets for each target separated according to the secondary target that should be avoided with higher priority. The given ranking error is equal to 1−AUC.

Mentions: In this section we show the results of the same experiments with the single-target activity profiles but with the alternative labeling from Table 3 applied to binary test sets (see Figures 10 and 11) and ranking test sets (see Figures 12 and 13). In this setup it was important to avoid one of the two secondary targets with higher priority than the other. The results show, that for the cytochrome P450 data set this ranking scheme is more beneficial regarding a binary classification than the one described in Table 1. Treating non-selective actives as decoys with a label of 0 seems to improve the ranking performance. The results of the ranking test sets show, that avoiding one secondary target with higher priority is more demanding for MT RANK and MC-SVM. However, the ranking SVM still shows a better performance in general. Changing the linear factors for the linear combinations to +3 for the main target, −2 for the secondary target to be avoided with higher priority, and −1 for the remaining secondary target did not show any significant improvement in performance for SVM LC.Figure 10


A ranking method for the concurrent learning of compounds with various activity profiles.

Dörr A, Rosenbaum L, Zell A - J Cheminform (2015)

Performance of the cytochrome P450s single-target data set with binary test sets. Each boxplot depicts the mean ranking error on the 20 randomly generated test sets for each target separated according to the secondary target that should be avoided with higher priority. The given ranking error is equal to 1−AUC.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4306736&req=5

Fig10: Performance of the cytochrome P450s single-target data set with binary test sets. Each boxplot depicts the mean ranking error on the 20 randomly generated test sets for each target separated according to the secondary target that should be avoided with higher priority. The given ranking error is equal to 1−AUC.
Mentions: In this section we show the results of the same experiments with the single-target activity profiles but with the alternative labeling from Table 3 applied to binary test sets (see Figures 10 and 11) and ranking test sets (see Figures 12 and 13). In this setup it was important to avoid one of the two secondary targets with higher priority than the other. The results show, that for the cytochrome P450 data set this ranking scheme is more beneficial regarding a binary classification than the one described in Table 1. Treating non-selective actives as decoys with a label of 0 seems to improve the ranking performance. The results of the ranking test sets show, that avoiding one secondary target with higher priority is more demanding for MT RANK and MC-SVM. However, the ranking SVM still shows a better performance in general. Changing the linear factors for the linear combinations to +3 for the main target, −2 for the secondary target to be avoided with higher priority, and −1 for the remaining secondary target did not show any significant improvement in performance for SVM LC.Figure 10

Bottom Line: The experiments show that ranking-based algorithms show an increased performance for single- and multi-target virtual screening.Moreover, compounds that do not completely fulfill the desired activity profile are still ranked higher than decoys or compounds with an entirely undesired profile, compared to other multi-target SVM methods.SVM-based ranking methods constitute a valuable approach for virtual screening in multi-target drug design.

View Article: PubMed Central - PubMed

Affiliation: Center for Bioinformatics Tübingen (ZBIT), University of Tuebingen, Sand 1, Tübingen, 72076 Germany.

ABSTRACT

Background: In this study, we present a SVM-based ranking algorithm for the concurrent learning of compounds with different activity profiles and their varying prioritization. To this end, a specific labeling of each compound was elaborated in order to infer virtual screening models against multiple targets. We compared the method with several state-of-the-art SVM classification techniques that are capable of inferring multi-target screening models on three chemical data sets (cytochrome P450s, dehydrogenases, and a trypsin-like protease data set) containing three different biological targets each.

Results: The experiments show that ranking-based algorithms show an increased performance for single- and multi-target virtual screening. Moreover, compounds that do not completely fulfill the desired activity profile are still ranked higher than decoys or compounds with an entirely undesired profile, compared to other multi-target SVM methods.

Conclusions: SVM-based ranking methods constitute a valuable approach for virtual screening in multi-target drug design. The utilization of such methods is most helpful when dealing with compounds with various activity profiles and the finding of many ligands with an already perfectly matching activity profile is not to be expected.

No MeSH data available.


Related in: MedlinePlus