How do you represent genomes?
Not just the genome, but genomeS. As Jan at Saaien Tist has mentioned, human (and other species) genomes are quiet variable. Though the linear representation of genome browsers makes perfect sense (like the UCSC Genome Browser, Ensembl, GBrowse and MapViewer among others) for much annotated data of the genome, structural variations are not so well visualized in a linear representation. And, as we are find the human and other specie genomes are quite variable, we might need to come up with another way to visualize these genomic data beyond the ‘reference genome’ linear model. Jan suggests deBruijn graphs,
pictured here. I find some difficulty in ‘visualizing’ how these are going to work for the _other_ annotations in the data. Though this representation looks like it might work great for CNV and the like, it seems to make viewing other types of data (expression, SNP, etc) more complicated. I’m looking forward to see how this develops.
Or perhaps we’ll be looking at genomes like this (ok, maybe not, but it’s geeky cool).