figshare
Browse
GRcalculator_11_20_16_pdf.pdf (2.09 MB)

GRcalculator: an online tool for calculating and mining drug response data

Download (2.09 MB)
poster
posted on 2016-11-21, 04:11 authored by Nicholas ClarkNicholas Clark, Marc Hafner, Michal Kouril, Jeremy Muhlich, Mario Niepel, Elizabeth Williams, Peter Sorger, Mario Medvedovic

Large-scale dose-response data and genomic datasets can be combined to discover novel drug-response biomarkers. There exist numerous online datasets of drug-response assays, but they are currently poorly accessible and their potential as a big data resource is limited due to lack of access and connection. Furthermore, it has recently been found that drug-response data often vary from one study to the next. A major reason for this variance is that traditional metrics of drug sensitivity such as IC50, Emax, and AUC values are confounded by the number of cell divisions taking place over the course of an assay. To solve this problem, we have developed GRcalculator, a suite of online tools found at www.grcalculator.org.  The tools use GR metrics (proposed recently by Hafner et al. in Nature), a set of alterative drug-response metrics based on growth rate inhibition that are robust to differences in nominal division rate and assay duration.

 

GRcalculator is a powerful, user-friendly and free tool for mining drug-response data using GR metrics. These metrics harmonize drug-response data, improving the discovery of novel drug-response biomarkers using big data as well as allowing for comparisons with patient-derived tumor cells that are generally slow growing. Direct access to LINCS drug-response datasets and, in the future, other public domain datasets is a unique functionality that will facilitate re-use of the valuable resources that these data represent. As well as mining datasets, the tool also offers calculation and visualization of GR metrics (and traditional metrics), generates publication-ready figures, and provides a unified platform for researchers analyzing drug sensitivity. For offline calculation and analysis, we have developed the GRmetrics R package (available via Bioconductor), which allows for use of larger datasets and inclusion of GR metrics calculations within existing R analysis pipelines.

Funding

U54HL127624

History