Uncovering a cover

November 5, 2011

Fig. 1: The EMBO Journal cover, showing a comparison of four genomes, which makes use of a publicly available bioinformatic visual analytic tool called Circos.

This is the cover of the European Molecular Biology Organization Journal (short: EMBO Journal) (Volume 28, Number 9
06 May 2009) (Fig. 1) is a nice example how scientific results can be made visible in a beautiful way. The freely available Circos tool was used to compare the also freely available genomes of human, chimpanzee, mouse, and zebrafish. Circos is an aesthetic appealing a visually relatively easy way of relating data sets to one another. In the times of -omics it can be handy for disciplines such as biochemistry, molecular biology, and genetics to present their data in a recognizable and interpretable manner.

An example of how Circos figures are used in scientific and popular literature in general is pictured below (Fig. 2). It gives a first idea about the impact that the visually appealing presentation of your data can have on general public readers and fellow scientists. PR of your science is getting more and more important I would say.

Fig. 2: Examples of Circos usage in scientific and popular literature as well as advertisement (originally from the Circos homepage).

In the following I would like to explain how these magnificent circular figures have to be interpreted based on the orignal publication of Martin I Krzywinski et al. in Genome Research, Vol 19, 1639-1645, 2009 where the author describes the general properties of the program and gives more examples of the programs capabilities. In order to explain the cover which is pictured above in Fig. 1, below in Fig. 3 you see the complete and more “clean” version of it.

Fig. 3: Complete visual genomic comparison, which is also displayed as a journal cover in Fig. 1. For a bigger picture size click the figure (extracted from a Circos visual guide which can be found here).

Fig. 4: Legend for Fig. 3. Click to enlarge (extracted from a Circos visual guide which can be found here).

When concentrating on the EMBO figure above, one thing is immediately apparent to the eye: The enormous amount of DNA similarity linkages between human (top right), zebrafish (top left), mouse (bottom left), and chimpanzee (bottom right) genomes. However, it is important to firstly note that this figure does NOT picture ALL possible similarity linkages between the genomes, but instead only compares the human genome to the other three. Every human chromosome has a unique color-code which is visible at the outermost edge in the “human section” of Fig. 3 next to the chromosome number (1-22, and X). This color-code also makes the similarity lines, which span across the figure, more recognizable. Every locus on every human chromosome can thus be linked to one (or more) loci on each of the three other genomes.

The most other details (inner rings) are not as easy comparable to each other, since they depict individually SELECTED information PER genome. As explained in Fig. 4 this means that the yellow and blue zone is differing per genome. The intention seemed to be to display the capabilities of the program rather than making the genomes perfectly comparable for a number of items like for example number of exons on that chromosome or types/names of genes. Instead each genome covers some unique information which can be traced back in the legend (Fig. 4).

It is thus obvious from this short summary of just one publicly available example what the advantages and drawbacks of circular gene comparison are. First of all it is appealing to the eye and with the help of a legend easily and quickly interpreted. Dealing with gene annotations actually becomes fun. However, it is also clear that this display technique rises and falls with its resolution. Of course it is impressive to compare four genomes almost literally in one blink of an eye, but the information content which can be extracted after studying the picture remains quite low. Although this is the case, in his paper Martin Krzywinski presents also ways of using Circos in a higher resolution manner. Instead of comparing four whole genomes to each other it is also possible to concentrate on one single “cancer” gene, which appears in four cancer types. Conserved mutations and recombinations become immediately visible. Circos thus, on a small-scale, is ideal to for example assist in tackling huge challenges like the interpretation of the cancer genome which is a new approach to subdivide cancer into genetic types instead of morphological ones. In the future this might allow more efficient therapies for similar typed cancers.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: