Section author: Gavin Huttley
This is just about the simplest possible Cogent script for evolutionary modelling. We use a canned nucleotide substitution model (the HKY85 model) on just three primate species. As there is only unrooted tree possible, the sequence names are all that’s required to make the tree.
>>> from cogent.evolve.models import HKY85
>>> from cogent import LoadSeqs, LoadTree
>>> model = HKY85()
>>> aln = LoadSeqs("data/primate_cdx2_promoter.fasta")
>>> tree = LoadTree(tip_names=aln.Names)
>>> lf = model.makeLikelihoodFunction(tree)
>>> lf.setAlignment(aln)
>>> lf.optimise(show_progress = False)
>>> print lf
Likelihood Function Table
======
kappa
------
5.9589
------
===========================
edge parent length
---------------------------
human root 0.0040
macaque root 0.0384
chimp root 0.0061
---------------------------
===============
motif mprobs
---------------
T 0.2552
C 0.2581
A 0.2439
G 0.2428
---------------