Home | Screening data | Screen comparisons | Search for compounds | Structure search

Chemical Genomics Matrix Data

This webportal allows navigation of the data generated for "Prediction of Synergism from Chemical-Genetic Interactions by Machine Learning" (Wildenhain et al. Cell Systems, 2015). Screening data can be accessed via the 'Screening data' and 'Screen comparison' tabs. Data for single molecules can be retrieved on the 'Search for compounds' and 'Structure search' tabs. The SONAR graph analysis is available via the compound pages in the right top corner (Example). To generate drug combination checkerboards you can use the shiny CheckerboardR.

Supplementary tables provided with the manuscript:

Table S1 Overview on compound libraries, unique molecules, unique strain identifiers and cryptagens in study.

Table S2 Unique compound identifiers and CGM identifiers for all compounds used in study.

Table S3 Selected sentinel strains and biological description.

Table S4 Naive Bayes AUC values for each sentinel strains in this study.

Table S5 Compound properties, information and experimental data for the 128 matrix molecules.

Table S6 Literature curation of yeast targets from literature compared with SONAR performance.

Table S7 Synergy prediction results versus the Bliss independence values measured for different datasets.

Table S8 Confusion matrices on test data for Spectrum, Maybridge Hitskit and ChemDiv Kinase libraries.

Table S9 Details for compounds used for Figures 7 & 8 with abbreviations and checkerboard results for this study.

Service provided by the Mike Tyers Laboratory