See: http://bioinformatics.erc.monash.edu/home/sarcher/ggscaleRanges/ggscaleRanges_eg.html

TODO:

• Make anchor-points more flexible and robust. Right now it is easy to mess up the plot by choosing xlims that are not exactly equal to the flanking anchorpoints.
• Merge graduated scale tranformations with function-of-length scale transformations
• Plot of actual magnification
• Add a ChIP-seq example to the .rmd