Selectivity data: assessment, predictions, concordance, and implications.

Could high-quality in silico predictions in drug discovery eventually replace part or most of experimental testing? To evaluate the agreement of selectivity data from different experimental or predictive sources, we introduce the new metric concordance minimum significant ratio (cMSR). Empowered by cMSR, we find the overall level of agreement between predicted and experimental data to be comparable to that found between experimental results from different sources. However, for molecules that are either highly selective or potent, the concordance between different experimental sources is significantly higher than the concordance between experimental and predicted values. We also show that computational models built from one data set are less predictive for other data sources and highlight the importance of bias correction for assessing selectivity data. Finally, we show that small-molecule target space relationships derived from different data sources and predictive models share overall similarity but can significantly differ in details.

[1]  Michal Vieth,et al.  What general conclusions can we draw from kinase profiling data sets? , 2013, Biochimica et biophysica acta.

[2]  Maria F. Sassano,et al.  Automated design of ligands to polypharmacological profiles , 2012, Nature.

[3]  Eric J. Martin,et al.  Profile-QSAR and Surrogate AutoShim Protein-Family Modeling of Proteases , 2012, J. Chem. Inf. Model..

[4]  Michael J. Keiser,et al.  Large Scale Prediction and Testing of Drug Activity on Side-Effect Targets , 2012, Nature.

[5]  Robert P. Sheridan,et al.  Three Useful Dimensions for Domain Applicability in QSAR Models Using Random Forest , 2012, J. Chem. Inf. Model..

[6]  Mindy I. Davis,et al.  Comprehensive analysis of kinase inhibitor selectivity , 2011, Nature Biotechnology.

[7]  George Karypis,et al.  Improved machine learning models for predicting selective compounds , 2011, BCB '11.

[8]  Eric J. Martin,et al.  Profile-QSAR: A Novel meta-QSAR Method that Combines Activities across the Kinase Family To Accurately Predict Affinity, Selectivity, and Cellular Activity , 2011, J. Chem. Inf. Model..

[9]  P. Hajduk,et al.  Navigating the kinome. , 2011, Nature chemical biology.

[10]  Michal Vieth,et al.  Structure-guided expansion of kinase fragment libraries driven by support vector machine models. , 2010, Biochimica et biophysica acta.

[11]  Stephen V Frye,et al.  The art of the chemical probe. , 2010, Nature chemical biology.

[12]  Ian A. Watson,et al.  Kinase inhibitor data modeling and de novo inhibitor design with fragment approaches. , 2009, Journal of medicinal chemistry.

[13]  Gavin Harper,et al.  Assessment of chemical coverage of kinome space and its implications for kinase drug discovery. , 2008, Journal of medicinal chemistry.

[14]  Ian A. Watson,et al.  Chemical fragments as foundations for understanding target space and activity prediction. , 2008, Journal of medicinal chemistry.

[15]  Jean-Pierre Doucet,et al.  Nonlinear SVM Approaches to QSPR/QSAR Studies and Drug Design , 2007 .

[16]  Michael L. Creech,et al.  Integration of biological networks and gene expression data using Cytoscape , 2007, Nature Protocols.

[17]  P. Clemons,et al.  Chemogenomic data analysis: prediction of small-molecule targets and the advent of biological fingerprint. , 2007, Combinatorial chemistry & high throughput screening.

[18]  Andreas Bender,et al.  "Bayes Affinity Fingerprints" Improve Retrieval Rates in Virtual Screening and Define Orthogonal Bioactivity Space: When Are Multitarget Drugs a Feasible Concept? , 2006, J. Chem. Inf. Model..

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

[20]  Kim E. Garbison,et al.  The Minimum Significant Ratio: A Statistical Parameter to Characterize the Reproducibility of Potency Estimates from Concentration-Response Assays and Estimation by Replicate-Experiment Studies , 2006, Journal of biomolecular screening.

[21]  M. Vieth,et al.  Kinomics-structural biology and chemogenomics of kinase inhibitors and targets. , 2004, Biochimica et biophysica acta.

[22]  Dragos Horvath,et al.  Predicting ADME properties and side effects: the BioPrint approach. , 2003, Current opinion in drug discovery & development.

[23]  T. Hunter,et al.  The Protein Kinase Complement of the Human Genome , 2002, Science.

[24]  Thorsten Joachims,et al.  Making large-scale support vector machine learning practical , 1999 .

[25]  J. H. Ward.  Hierarchical Grouping to Optimize an Objective Function , 1963 .