Growth of Ligand-Target Interaction Data in ChEMBL Is Associated with Increasing and Activity Measurement-Dependent Compound Promiscuity

Compounds with high-confidence target annotations and activity measurements in the original and current release of the ChEMBL database have been compared to better understand how the growth of compound activity data might influence the spectrum of ligand-target interactions and the degree of target promiscuity among active compounds. Compared to the original ChEMBL release, a significant increase in the proportion of target promiscuous compounds was observed in the current version. The presence of these compounds led to large-magnitude changes in compound activity-based target and target family relationships and to a reorganization of major target communities. Surprisingly, however, this strong trend toward increasing target promiscuity was largely caused by growth of compounds with exclusive IC(50) measurements. By contrast, compounds with available equilibrium constants, which were also added in large amounts, did not substantially alter compound-based target relationships and notably contribute to increasing target promiscuity. These findings suggest that apparent compound promiscuity is much dependent on experimental conditions under which activities are determined and that care should be taken when evaluating promiscuity and polypharmacology on the basis of assay-dependent activity measurements.

[1]  G. Bemis,et al.  The properties of known drugs. 1. Molecular frameworks. , 1996, Journal of medicinal chemistry.

[2]  Yong-Jin Xu,et al.  Using Molecular Equivalence Numbers to Visually Explore Structural Features that Distinguish Chemical Libraries. , 2002 .

[3]  P. Shannon,et al.  Cytoscape: a software environment for integrated models of biomolecular interaction networks. , 2003, Genome research.

[4]  G. V. Paolini,et al.  Global mapping of pharmacological space , 2006, Nature Biotechnology.

[5]  Michael J. Keiser,et al.  Relating protein pharmacology by ligand chemistry , 2007, Nature Biotechnology.

[6]  R. Solé,et al.  Data completeness—the Achilles heel of drug-target networks , 2008, Nature Biotechnology.

[7]  A. Hopkins Network pharmacology: the next paradigm in drug discovery. , 2008, Nature chemical biology.

[8]  J. Bajorath,et al.  Systematic computational analysis of structure-activity relationships: concepts, challenges and recent advances. , 2009, Future medicinal chemistry.

[9]  Andreas Bender,et al.  Databases: Compound bioactivities go public , 2010 .

[10]  Jürgen Bajorath,et al.  Polypharmacology Directed Compound Data Mining: Identification of Promiscuous Chemotypes with Different Activity Profiles and Comparison to Approved Drugs , 2010, J. Chem. Inf. Model..

[11]  J. Bajorath,et al.  Systematic analysis of public domain compound potency data identifies selective molecular scaffolds across druggable target families. , 2010, Journal of medicinal chemistry.

[12]  J. Bajorath,et al.  Activity landscape representations for structure-activity relationship analysis. , 2010, Journal of medicinal chemistry.

[13]  J. Bajorath,et al.  BindingDB and ChEMBL: online compound databases for drug discovery , 2011, Expert opinion on drug discovery.

[14]  Pekka Tiikkainen,et al.  Analysis of Commercial and Public Bioactivity Databases , 2012, J. Chem. Inf. Model..

[15]  Sean Ekins,et al.  Towards a gold standard: regarding quality in public domain chemistry databases and approaches to improving the situation. , 2012, Drug discovery today.

[16]  John P. Overington,et al.  ChEMBL: a large-scale bioactivity database for drug discovery , 2011, Nucleic Acids Res..

[17]  M. Gilson,et al.  Public domain databases for medicinal chemistry. , 2012, Journal of medicinal chemistry.

[18]  Huimin Zhao,et al.  Customized optimization of metabolic pathways by combinatorial transcriptional engineering , 2012, Nucleic acids research.