david's really interesting pages | Just another WordPress site

Liam Revell; Phylogenetic uncertainty

I really like the idea of deigning visuals that transport information peripherally, and so its not surprising that I think Liam Nevell is onto something really cool here. The likelihood of a phylogenetic position is reflected by the color value surrounding each branch. Very cool.

There are 2 Comments to this article

Christopher Lutz says:
02/20/2013

Hmm … I’m not really convinced. Maybe the branches should be drawn further apart. As it is now, there is too much of an overlap of the branches. Maybe one can see that there is a higher probability for the branches in the upper right corner, but it is nearly impossible to tell which of the individual branches actually are more probable.

Maybe instead of using colors, why not just make the more probable branches thicker than the less likely branches ?
Or – if you want to stick to coloring – make the more likely branches blueish and the less likely branches yellowish. In either case, the overlapping of branches should be avoided.

d maas says:
02/20/2013

Of course this is really raw. It would profit greatly from radial ending instead of that linear cutoff, or a vector-based thickness / color falloff, as you suggest. Might even be a case for diffusion curves. What I like is that his is being generated from the data, and not as a manual ‘artistic’ driven layer.

There are older hand-drawn graphs that I think are fantastic references for possible visual treatments on this. I’ll have to make a post about this…

Write a Comment