In today’s tip I’d like to introduce you to the Cancer Genome Workbench, or CGWB. The workbench gathers cancer information from a wide variety of projects including Johns Hopkins University and GlaxoSmithKline Cancer Cell Line Genomic Profiling Data, NCI’s Therapeutically Applicable Research to Generate Effective Treatment (TARGET), NHGRI’s Tumor Sequencing Project (TSP), The Cancer Genome Atlas (TCGA), and the Sanger Center’s COSMIC initiative and presents the cumulative data as high-level summary visualizations. The CGWB’s genome-browser view is built on a UCSC Genome Browser backbone, for power and flexibility.
I noticed an announcement in the May 7th Nature Signaling Gateway Update email that the NCI-Nature Pathway Interaction Database – May Update was featuring a bioinformatics primer on The Cancer Genome Workbench. The primer is great & goes into much more detail about the Cancer Genome Workbench than I will be able to in this quick tip. I strongly check the primer, and the workbench out. When I went over to the workbench to explore, I quite honestly was a bit taken back by the complexity of the displays – the amount of data presented in their summary visualizations are somewhat intense.
I hope that in my tip movie I will be able to convince you that the small investment you will need to do to get acclimated to their images is well worth the amount of data you will quickly understand how to analyze. The views are so data rich, it takes a bit of adjusting to – there is very little labeling (to keep displays as clean as possible) and information is provided via pop-up messages as you scroll over the display. Once I got past the intensity of the displays, I was really amazed by the scope of data visualized in CGWB displays – data on every chromosome & gene over multiple datasets/experiments, in one 2D image. As the NCI primer says, cancer is complex – really complex. Being able to see such ‘big picture’ views as those provided by the Cancer Genome Workbench is a really powerful analysis aid. I for one am impressed with this resource, which is why I’ve chosen to feature it today.
In my 5 minute tip I was only able to show you the briefest of glimpses of the CGWB landscape and heatmap views. I was not able to show you the details of wither view, including a hyperlinked list of genes with the highest mutation frequencies. Nor was I able to show you the full scope of other views which include genome browser views (based on the UCSC Genome browser, as I mentioned earlier), correlation plots, protein domain views, 3D vizualizations, as well as next-gen and trace sequence views. Check out figure 1 of the bioinformatics primer to see examples of those.
I’ve added a citation to the original CGWB publication. It was published in 2007, and so does not cover all the current functions of the workbench, but I think reading it might help give you an idea of the workbench because it goes into the goals and background that the CGWB is based on more than the primer, which is much more up-to-date and focuses on the functionality of the workbench. In this paper you can also read how the authors utilized the workbench to analyze three public datasets, and see how it expanded their research findings.
All & all, I think the Cancer Genome Workbench is an amazing resource for cancer research. Be sure to check out the tip movie, the primer, the original CGWB publication and especially the CGWB! Thanks for joining us for this week’s tip.
Zhang, J., Finney, R., Rowe, W., Edmonson, M., Yang, S., Dracheva, T., Jen, J., Struewing, J., & Buetow, K. (2007). Systematic analysis of genetic alterations in tumors using Cancer Genome WorkBench (CGWB) Genome Research, 17 (7), 1111-1117 DOI: 10.1101/gr.5963407