Reproducibility of Chemopredictors from Cell Lines

This web page contains files summarizing the analyses of data described in "Deriving Chemosensitivity from Cell Lines: Forensic Bioinformatics and Reproducible Research in High-Throughput Biology" by Keith A. Baggerly and Kevin R. Coombes. The paper has now appeared in the Annals of Applied Statistics, 2009 3(4):1309-1334. All analyses were performed by Keith A. Baggerly and Kevin R. Coombes.

This page was updated on Sunday, May 9, 2010, to include the paper reference above. Everything below the following horizontal line is frozen, and will not change. This page was updated on Thursday, November 19, 2009, after galleys were marked up. We have made one change since Thursday, September 24, 2009, adding a dashed line to one figure in examiningDoxorubicinInDetail and rebundling the related zip files. Our intention is to avoid changing the files posted here after publication, allowing the web site to serve as permanent documentation of our analysis. Any changes will be posted on a separate page designed for addenda, errata, corrigenda and other related latinisms.

Here is a list of the available material. The supplementary reports are in PDF format. The source code, intermediate results, and public data sources are in zip files, starting from a common directory. The simplest approach is to download "Everything" and unzip the file. For a minimal reconstruction, we download the Source Code (containing Scripts and RNowebSource), Raw Data, and Matlab Files, create folders for Figures, Latex, RDataObjects, and RTangle, and source "Scripts/runAll.R". Since runAll.R is system specific, some modification will be required to point to your local installations of pdflatex, etc. This creates all of the Figures, Reports, RDataObjects, and R scripts (in RTangle) used to produce the results. This takes about 7.5 minutes on my MacBook Pro.

• Everything (177 MB)
• Source Code (122 KB)
• RawData (121 MB)
• MatlabFiles (32.7 MB)
• Selected Supplementary reports (others are in "Latex" below):
  • examiningDoxorubicinInDetail (1.8 MB)
  • matchCisplatinHeatmap.pdf (454 KB)
  • matchPemetrexedHeatmap.pdf (1.1 MB)
  • Supplementary reports involving combination therapy.
  • checkingCellLines.pdf (719 KB) examines drug sensitivity values for the cell lines used.
  • enumeratingCellLines.pdf (283 KB) enumerates which cell lines are sensitive and resistant to assorted drugs according to disparate information sources.
  • examiningDocetaxelInDetail.pdf (1.2 MB)
• Intermediate results:
  • Figures (969 KB)
  • Latex (2.7 MB)
  • RDataObjects (26.3 MB)
  • RTangle (33 KB)
• PaperFigures (415 KB) scripts showing how the figures used in the paper are assembled, and the figures themselves.

Our analysis source code relies on a number of software programs and auxiliary packages; we provide scripts, not stand-alone executables. Detailed descriptions of the packages (with version numbers) are listed in the methods section of the main manuscript. The pieces of software required to execute the source code can be obtained from the following locations:

• The main program and several package libraries from the R environment for statistical programming.
• Several package libraries from BioConductor.
• A working version of LaTeX, the text formatting system based on the work of Donald Knuth.
• The commercial softare program MATLAB