Ideas for graphing data?

[puggie]

I have developed a method that identifies some genomic features and does so tissue-specifically. Now I would like to make some visual representation of this. E.g in one type of tissue I get lets say 1000 hits of varying intensity across the genome chromosomes. In other tissues I could get lets say 500-1500 hits of varying intensity across chromosomes.

It can be boiled down to a genomic coordinate accompanied by a intensity/frequency score. The coordinate is not super-important for the graphical representation, hence no need to depict the exact chromosome, however it would aid interpretation I guess.

Is there some 'fancy' way of doing this besides a heatmap or the like for every chromosome?

All inputs appreciated.

[blancha]

Circos plots?

[DPCook]

Circos plots?

I second this. Circos plot with histogram tracks for each tissue could look really good. The only problem I can see would be resolution. If you want to show differences within small ranges (eg. Differences in histone marks within a 100kb range), this may not be best. Because the plot represents the whole genome, a single pixel often represents ~1Mb. However if you can bin your values into ~1Mb windows, you're golden.

http://circos.ca/

[Richard Finney]

Some guy named Ying Hu wrote some Circos thing in R (Who?)

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3921174/

OmicCircos: A Simple-to-Use R Package for the Circular Visualization of Multidimensional Omics Data