KML trees

I managed to knock together a NEXUS file for 9 species from the fusion table we created:


#NEXUS


[tree file with latitude and longitude appended to the taxon name]


BEGIN TREES;


tree * Oryzomys = (('O.acritus lat=-14.71long=-61.03'),
(('O.andersoni Alat=-17.40long=-59.51'),
(('O.curasoe lat=12.17long=-69.00'),
((((('O.emmonsa lat=-3.65long=-52.37'),
('O.maracajuensis lat=-21.63long=-55.15')),
('O.marinhus lat=-14.67long=-45.83')),
('O.scotti lat=-15.90long=-48.80')),
('O.seuanezi lat=-22.42long=-42.03')),
((('O.tate lat=-1.42long=-78.20')))))));


END;


I then converted this into a KML file using this nifty tool courtesy of iphylo.org
Opening this KML file in Google Earth then displays the tree superimposed onto the map.
Since this was a little experiment to see if I could get it to run the tree is not correct and needs some tweaking.

R me hearties

One of the things we wanted to do with our data was create a tree of the species we have.
I immediately thought of using R, mainly because I took a course in basic R last semester and I keep hearing people sing it's praises.  Another reason for using R is that if you want to do something, chances are someone has already done it and posted the code for you.

I found a great example of creating phylogenies in R here.  The radial tree example below is what I'm aiming for although installing the right package has proved to be a little harder than expected.

At first glance it seemed as though there was a lot of stuff relating to R and creating phylogenies, however I soon discovered that this wasn't so.  Many of the R tutorials and examples do not help in building a tree from the data you have.
So the next task is to find a way to build the tree then play with it R.

Phylo Table Assemble!

Phew!  It's been a while since I've updated this, partly due to the environmental stats hell I just dragged myself out of.


Anyway.
My friend Peggy has been stalwartly manipulating the data we captured and has been experimenting with ways to use the data, for now I'll be catching up with what we've done in the last couple of weeks.


After we managed to extract all that juicy data from the source paper and some minor data cleaning, we had an Excel file containing each of the 341 species and all the information relating to each one. 
This was then imported into Google Docs, giving us more ways to play around with this data using some of the techniques we'd seen during the Phyloinformatics project.


The first we tried (because it looks cool) was to use display the results on Google maps:

Zooming in on South America you can clearly see the species displayed as green dots.  


Clicking on one of these dots displays the information in a speech bubble.










So far so good, but there's more that can be done with this data than just using the coordinates.  We still have data on the family and orders of species and the locality of the discovery i.e. Continental (C), insular (I) or marine (M).


To visualise this locality and get a grasp of where the majority of discoveries where we tried to make a chart, in this case a pie chart.
Peggy worked out that you first had to aggregate the data by the locality, then simply hit the pie chart button and...

..freshly baked pie chart showing that the majority of species discovered between 1992 - 2000 were found in continental areas.










This is all quite basic stuff, so we're looking at doing something a little more challenging.
More to come.