I generated data that compared DNA sequences from different samples with their counterparts in chosen references. I produced a network to show an overview of the relationships. But I wanted inspect one particular subcluster.
In particular, I wanted to to know how similar or different my samples were to the references. I gathered my data in a table that looked like this:
where each column is actually split into more columns, with fields that may or may not be filled.
But I'm just a flawed human being and not a machine, and this table gave me no insight whatsoever. So I decided to present my comparison in the form of a stacked alignment. My expectation was something like this:
But there were too many elements in my subnetwork, that what I got was a this headache:
where, as you can see, you cannot see anything. The top and the bottom are far from each other and difficult to compare. Also, I have to choose only one reference top put to the top.
But with larger alignments, it was of no help at all.
So I chose to present my data via circos. This way, all sequences are of equal distance to each other, and I don't have to choose just one reference.
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Circos was created by Martin Krzywinski, and you can read more about it at http://circos.ca/.