A while ago we published a review about multidimensional cancer genomics data visualization in Genome Medicine. There, we focused on effective and common visualization techniques for exploring oncogenomics data and we discussed a selection of tools that allow researchers to effectively visualize multidimensional oncogenomics datasets. Since our research field is constantly evolving we thought we could share an update of the tools and links.
Please note that the portals and tools listed in the table below were selected because they allow to visualize several dimensions of cancer genomics data together. The list thus does not pretend to be a comprehensive set of resources for cancer research. Multiple dimensions in cancer genomics data are different variables measured per tumor sample and genomic elements (for example, the genes), such as copy number alterations (CNAs), mutations, gene expression changes and methylation changes.
Updates from the publication of the review:
We have recently learnt about two new interesting tools for multidimensional cancer genomics data visualization that are now included in the updated table in this post.
Cancer Landscapes, which uses network visualization to represent associations between variables (such as mRNA expression, methylation or copy number alterations). The associations are computed using high performance optimization methods.
Next Generation Clustered Heatmaps, which uses matrix heatmaps to visualize genomic and transcriptomic data from tumors. The tool allows to zoom in and out, and pan across the heatmap to see details at many levels of resolution. Other interactive controls enable searching for specific heatmap entries, generating production quality PDFs, and linking out to information related to rows, columns, and individual heatmap entries. Data from TCGA is ready to be explored with this tool.
Another update we would like to mention is the brand new version of Gitools, (v.2.0), which is available as a preview version. This new version contains substantial changes, among which we would like to highlight that, 1) it is now able to load very large data matrices in a memory efficient way; 2) it possesses an improved user’s interface; and 3) it includes new functions specifically designed for the study of multidimensional cancer genomics data.
Table of tools listed in alphabetical order
We distinguish between three main approaches commonly used to represent multidimensional oncogenomics data: genomic coordinates, heatmaps and networks. The tools in the table are classified according to the visualization type they use.
|Name||Visualization type||Tool type|
|cBio Cancer Genomics Portal||Networks – Matrix Heatmaps||Webtool|
|CircleMap||Circular Heatmaps||Command line application – Webtool|
|Circos||Circular Genomic Coordinates||Command Line Application|
|Caleydo StratomeX||Matrix Heatmap with option to visualize pathway maps||Desktop Application (Java)|
|Cytoscape||Networks||Desktop Application (Java)|
|Genomica||Matrix Heatmap – Genomic coordinates||Desktop Application (Java)|
|Gitools||Matrix Heatmap with interactive features||Desktop Application (Java)|
|Integrative Genomics Viewer (IGV)||Genomic Coordinates||Desktop Application (Java)|
|IntOGen||Matrix Heatmaps with interactive features||Webtool|
|NAViGaTOR||Networks||Desktop Application (Java)|
|Next Generation Clustered Heatmaps||Heatmaps||Webtool|
|Regulome Explorer||Circular and linear Genomic coordinates – Networks||Webtool|
|Savant Genome Browser||Genomic Coordinates||Desktop Application (Java)|
|The Cancer Genome Workbench||Genomic Coordinates – Heatmap||Webtool|
|UCSC Cancer Genomics Browser||Genomic Coordinates – Heatmaps||Webtool|
How to visualize multidimensional cancer genomics data?
Nice compilation. Another relevant tool not mentioned above is IGB, the Integrated Genome Browser: bioviz.org/igb. It has some nice features and optimizations that allow it to load and smoothly navigate over large, whole-chromosome data sets (and not just chr22 ;).
The Integrated Genome Browser: free software for distribution and exploration of genome-scale datasets. Disclaimer: I’m a former IGB developer.
Thanks very much for the suggestion Steve.