I’ve always been a fan of clever graphical displays that convey key data points. They can be so effective when done well. And from way back when I was first exposed to the logo-style histogram displays for conserved promoter sequences they’ve always struck me as particularly suited to this method. Usually for ease-of-use I’ve gone to WebLogo. It’s easy, and quick to use. But recently I learned of another tool that provides the logos and has a bit more to it. The iceLogo tool seems to be a useful option for these now as well, particularly suited to protein motif display.
A brief correspondence in a recent issue of Nature Methods introduced me to iceLogo. When I first tried it out, I had to download it and run it locally (and you can still do that). But there is also a web interface for this tool now, and that will be the focus of my tip of the week.
You can enter a set of sequences as a multiple sequence alignment, then you can choose the reference set of everything in Swiss-Prot or set your own background, and then visualize the sequence data with conserved positions in a couple of ways. You can see the larger letter sizes using several logo styles, or you can access a heat-map style display. You can download the images (although I often just screen shot things like this). I wish it had a clear button to easily swap out my items of interest, but that’s not a huge problem really. In this tip of the week movie I run their sample sequence to show you how it displays the data.
Their manual (PDF) goes in to the detail on the statistics and the theory. The Nature Methods paper is quite short–there real meat is in the manual. And there’s actually a neat feature to the manual PDF: you can click on Figure 1.4 to directly load up that example in the web interface and change it around if you want.
Check out iceLogo if you want to explore these types of visualizations for your data. Visit the iceLogo web interface here.
Just a final note: there is also a SOAP server that you could access with other tools as a web service that might be handy if you are setting up analysis pipelines in various situations.
Colaert, N., Helsens, K., Martens, L., Vandekerckhove, J., & Gevaert, K. (2009). Improved visualization of protein consensus sequences by iceLogo Nature Methods, 6 (11), 786-787 DOI: 10.1038/nmeth1109-786
UPDATE: Niklaas from the iceLogo team came by with an update to the web site address–go here now for that. http://iomics.ugent.be/icelogoserver/main.html