Limits...
The 'SAR Matrix' method and its extensions for applications in medicinal chemistry and chemogenomics.

Gupta-Ostermann D, Bajorath J - F1000Res (2014)

Bottom Line: The SARM method has originally been designed for the extraction, organization, and visualization of compound series and associated SAR information from compound data sets.It has been further developed and adapted for other applications including compound design, activity prediction, library extension, and the navigation of multi-target activity spaces.The SARM approach and its extensions are presented here in context to introduce different types of applications and provide an example for the evolution of a computational methodology in pharmaceutical research.

View Article: PubMed Central - PubMed

Affiliation: Department of Life Science Informatics, B-IT, LIMES Program Unit Chemical Biology and Medicinal Chemistry, Rheinische Friedrich-Wilhelms-Universität, Bonn, D-53113, Germany.

ABSTRACT
We describe the 'Structure-Activity Relationship (SAR) Matrix' (SARM) methodology that is based upon a special two-step application of the matched molecular pair (MMP) formalism. The SARM method has originally been designed for the extraction, organization, and visualization of compound series and associated SAR information from compound data sets. It has been further developed and adapted for other applications including compound design, activity prediction, library extension, and the navigation of multi-target activity spaces. The SARM approach and its extensions are presented here in context to introduce different types of applications and provide an example for the evolution of a computational methodology in pharmaceutical research.

No MeSH data available.


Neighborhood-based potency prediction.An NBH of virtual compound X is marked in blue in a model SARM and displayed in detail. Compounds E and G share the same substituents and core with X, respectively, and the third neighbor D combines the core and substituent of E and G, respectively. At the bottom, the equation to predict the potency of X from the potency values of E, G, and D is shown.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4215758&req=5

f5: Neighborhood-based potency prediction.An NBH of virtual compound X is marked in blue in a model SARM and displayed in detail. Compounds E and G share the same substituents and core with X, respectively, and the third neighbor D combines the core and substituent of E and G, respectively. At the bottom, the equation to predict the potency of X from the potency values of E, G, and D is shown.

Mentions: VCs contained in SARMs provide immediate suggestions for compound design. Because VCs represent unexplored key-value combinations derived from data set compounds, the union of VCs from all SARMs provides a “chemical space envelope” for a given compound set or library. VCs originating from SAR-informative matrices represent natural focal points for interactive compound design. Moreover, the potency of many virtual compounds can be predicted by applying a compound neighborhood (NBH) principle5, as illustrated inFigure 5. An NBH of a given VC is defined by three adjacent real compounds that contain the core of the VC (compound G inFigure 5), its substituent (compound E) and the core and substituent of G and E (compound D). The potency of the VC can then be predicted by applying the additivity assumption underlying Free-Wilson analysis9 using the simple equation shown inFigure 5. The putative potency value of the VC results from the sum of (logarithmic) potencies of the two real compounds sharing the same core and substituent with the VC, respectively, minus the potency of the compound that contains the core structure and substituent of the two other real compounds. Thus, from NBHs, “mini-QSAR” models are derived for activity prediction. For each candidate VC, qualifying NBHs are collected across all SARMs, individual potency predictions are carried out, and their consistency is evaluated, for example, by calculating standard deviations for predictions5. In benchmark calculations on six different sets of G protein-coupled receptor ligands, potency values of subsets of test compounds falling into continuous local SAR regions were accurately predicted using the NBH-based approach, and prediction accuracy generally increased with the number of qualifying NBHs5. This is also relevant for practical applications. For potency prediction, candidate VCs should be prioritized for which multiple NBHs are available. For example, for the set of 509 purinergic receptor ligands (vide supra), 5167 of 17,445 VCs were found to have at least three qualifying NBHs. Hence, in these cases, the consistency of potency predictions can be assessed. Such candidate VCs can be explored in a systematic manner. For libraries tested in individual assays, VCs predicted to be consistently active on the basis of multiple NBHs provide preferred candidates for target/assay-dependent library expansion and focusing.


The 'SAR Matrix' method and its extensions for applications in medicinal chemistry and chemogenomics.

Gupta-Ostermann D, Bajorath J - F1000Res (2014)

Neighborhood-based potency prediction.An NBH of virtual compound X is marked in blue in a model SARM and displayed in detail. Compounds E and G share the same substituents and core with X, respectively, and the third neighbor D combines the core and substituent of E and G, respectively. At the bottom, the equation to predict the potency of X from the potency values of E, G, and D is shown.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4215758&req=5

f5: Neighborhood-based potency prediction.An NBH of virtual compound X is marked in blue in a model SARM and displayed in detail. Compounds E and G share the same substituents and core with X, respectively, and the third neighbor D combines the core and substituent of E and G, respectively. At the bottom, the equation to predict the potency of X from the potency values of E, G, and D is shown.
Mentions: VCs contained in SARMs provide immediate suggestions for compound design. Because VCs represent unexplored key-value combinations derived from data set compounds, the union of VCs from all SARMs provides a “chemical space envelope” for a given compound set or library. VCs originating from SAR-informative matrices represent natural focal points for interactive compound design. Moreover, the potency of many virtual compounds can be predicted by applying a compound neighborhood (NBH) principle5, as illustrated inFigure 5. An NBH of a given VC is defined by three adjacent real compounds that contain the core of the VC (compound G inFigure 5), its substituent (compound E) and the core and substituent of G and E (compound D). The potency of the VC can then be predicted by applying the additivity assumption underlying Free-Wilson analysis9 using the simple equation shown inFigure 5. The putative potency value of the VC results from the sum of (logarithmic) potencies of the two real compounds sharing the same core and substituent with the VC, respectively, minus the potency of the compound that contains the core structure and substituent of the two other real compounds. Thus, from NBHs, “mini-QSAR” models are derived for activity prediction. For each candidate VC, qualifying NBHs are collected across all SARMs, individual potency predictions are carried out, and their consistency is evaluated, for example, by calculating standard deviations for predictions5. In benchmark calculations on six different sets of G protein-coupled receptor ligands, potency values of subsets of test compounds falling into continuous local SAR regions were accurately predicted using the NBH-based approach, and prediction accuracy generally increased with the number of qualifying NBHs5. This is also relevant for practical applications. For potency prediction, candidate VCs should be prioritized for which multiple NBHs are available. For example, for the set of 509 purinergic receptor ligands (vide supra), 5167 of 17,445 VCs were found to have at least three qualifying NBHs. Hence, in these cases, the consistency of potency predictions can be assessed. Such candidate VCs can be explored in a systematic manner. For libraries tested in individual assays, VCs predicted to be consistently active on the basis of multiple NBHs provide preferred candidates for target/assay-dependent library expansion and focusing.

Bottom Line: The SARM method has originally been designed for the extraction, organization, and visualization of compound series and associated SAR information from compound data sets.It has been further developed and adapted for other applications including compound design, activity prediction, library extension, and the navigation of multi-target activity spaces.The SARM approach and its extensions are presented here in context to introduce different types of applications and provide an example for the evolution of a computational methodology in pharmaceutical research.

View Article: PubMed Central - PubMed

Affiliation: Department of Life Science Informatics, B-IT, LIMES Program Unit Chemical Biology and Medicinal Chemistry, Rheinische Friedrich-Wilhelms-Universität, Bonn, D-53113, Germany.

ABSTRACT
We describe the 'Structure-Activity Relationship (SAR) Matrix' (SARM) methodology that is based upon a special two-step application of the matched molecular pair (MMP) formalism. The SARM method has originally been designed for the extraction, organization, and visualization of compound series and associated SAR information from compound data sets. It has been further developed and adapted for other applications including compound design, activity prediction, library extension, and the navigation of multi-target activity spaces. The SARM approach and its extensions are presented here in context to introduce different types of applications and provide an example for the evolution of a computational methodology in pharmaceutical research.

No MeSH data available.