Check out the introduction slides to review the goals and context for the event! If you’re interested in learning more, links to relevant papers are included in the Useful Links section below. There are a couple ways that you can get started.
.Rmd files with RStudio or any other editor that you’re comfortable using.Free accounts on RStudio Cloud have unlimited access to cloud computing until August 3, 2020! Unfortunately, after that, free accounts will be limited to 15 hours per month.
After going over the introductory slides and presentation, work through these reinforcement questions to make sure you and your team has a good understanding of replicability and the problem.
analysis_template (Rmd) (html) : R markdown template which each team will use to create a fully reproducible analysis with the goal of assessing and interpreting the replicability of two pharmacogenomic experiments. This document will contain all of the text and code of their analyses, which are quided by a series of questions. The tools and concepts needed to answer the questions are explored in the tutorials.Data files are included under the data folder.
rawPharmacoData.rds : raw data file generated by downloadData.R that contains drug response data at every dose for each cell line and drug used in both studies.
summarizedPharmacoData.rds : summarized data file generated by downloadData.R that contains drug response data (combined over all doses) for each cell line and drug used in both studies.
modelSummarizedPharmacoData.rds : summarized data file generated in the supplement_dose_response.Rmd tutorial that contains recomputed drug response data (combined over all doses) for each cell line and drug used in both studies.
Tutorials are included under the tutorial folder.
Each tutorial contains text and code that explores various aspects of data science, replicability, and reproducibility. Tutorial 0a provides a gentle introduction to R for those with limited programming experience, and Tutorial 0b introduces some of the main functions in the dplyr and tidyr packages. Tutorials 1a and 1b get us started with exploring the two original datasets from the CCLE and GDSC studies, rawPharmacoData.rds and summarizedPharmacoData.rds. Tutorials 2a and 2b dig deeper into specific issues that can impact replicability and provide ideas for things to look into for this Replicathon. Finally, the Supplementary Tutorials provide more details that are useful but not immediately necessary for getting started.
1a (Rmd) (html) : “Exploring Pharmacological Data with the rawPharmacoData Dataset”
1b (Rmd) (html) : “Exploring Replicability with the summarizedPharmacoData Dataset”
2a (Rmd) (html) : “Digging Deeper with Cell Line and Drug Subgroups”
2b (Rmd) (html) : “Digging Deeper with Drug Response Summarization”
Supplementary Tutorials
For full reproducibility, this is the script to generate the RDS files. This step has already been done for you, and the data files should be available as part of this repository without running the script.
downloadData.R : a script that uses the PharmacoGx package to format two datasets (raw and summary level) and save them as RDS files.CCLE (Cancer Cell Line Encyclopedia) Study from March 2012
GDSC (Genomics of Drug Sensitivity in Cancer) Study from March 2012
Reanalysis of CCLE and GDSC from December 2013
Commentary 1 on the reanalysis from December 2016
Commentary 2 on the reanalysis from December 2016
Commentary 3 on the reanalysis from December 2016
Revisiting the reanalysis from August 2017
To ensure a safe, enjoyable, and friendly experience for everyone who participates, we have a strict code of conduct that all participants are expected to follow.