Enabling transparent and collaborative computational analysis of 12 tumor types within The Cancer Genome Atlas

Abstract

The Cancer Genome Atlas Pan-Cancer Analysis Working Group collaborated on the Synapse software platform to share and evolve data, results and methodologies while performing integrative analysis of molecular profiling data from 12 tumor types. The group's work serves as a pilot case study that provides (i) a template for future large collaborative studies; (ii) a system to support collaborative projects; and (iii) a public resource of highly curated data, results and automated systems for the evaluation of community-developed models.

Figure 1

Molecular profiling data sets in the Pan-Cancer project. Each circular plot displays the total number of samples analyzed across each of the 12 tumor types in the Pan-Cancer project. Samples are arranged in the same order in each concentric circle for each tumor type. Different circles are colored according to whether the sample was profiled using the most current platform, was profiled using a legacy platform or was not profiled. Each data set, including older versions, is available in Synapse (syn300013).

Figure 2

Schematic of the Pan-Cancer analysis workflow. Data were aggregated and standardized from the TCGA DCC, Broad Firehose and individual analysis working groups and processed into easy-to-use tab-delimited files. Collaborators used a variety of analytic tools, such as R, Python, Unix shell and the web client, to interact with data in Synapse while also storing results, provenance records, analysis descriptions and source code. For a subset of these results (for example, patient survival predictions), Synapse carried out automated performance evaluations and displayed results on a real-time leader board, which were available to collaborators to perform comparative meta-analysis or adapt model source code to additional applications.

Figure 3

Example provenance graph of a multistep workflow showing interaction between the analysis of three researchers. The provenance record consists of two types of nodes—activities (shown as red boxes above) performed by a researcher and input and output files of these actions (shown as file and folder icons and identified by their name and Synapse ID). In addition, every activity has metadata associated with it to further describe the details of the actions performed. This specific graph shows the workflow used to perform comparative analysis of two mutation-calling algorithms—MuSiC and MutSig. For MuSiC, the provenance of analysis is displayed from input data to derivation of mutation calls. Provenance records may be further expanded (ellipses) to trace the origin of input files to their original data source in Firehose, DCC or personal communications with AWG members. For brevity, the MutSig graph is not expanded. This graph was produced from version 2 of the data in doi:10.7303/ syn1750331.