phylogeny using illumina data

[elhb]

Hi,

I've got illumina data from a number of human samples that I know are related (hopefully distantly enough to have informative differences) and would like to build a phylogenetic tree using SNP and short INDEL calls etc from GATK, samtools, dindel or similar.

A program that can make use of loads both homozygous and heterozygous variants (and both substitutions and indels) preferably in vcf format would be great!

Ok there might not be such a program though ...
... could anyone point me in the right direction?

[mbblack]

While I have not actually done any phylogenetics in quite a long while, there are a number of tools you could use for this.

Some lists of phylogentic inference softeware:

http://en.wikipedia.org/wiki/List_of_phylogenetics_software

http://evolution.genetics.washington.edu/phylip/software.html

But it is not like you are just going to be able to load up your vcf files, click a few options and go.

One way to do it would be to prepare a matrix of you features (indels or whatever) and then score each sample for presence or absence and use some distance based Bayesian or likelihood methods. But you are going to have to extract suitable information from your SNP and indel calls for phylogentic inference.

Bayesian methods seem popular with SNP data (Mrbayes seems to get cited often in my admittedly limited reading of this kind of work), so a quick literature search should point you to what others have or are using for phylogentic inference with SNPs.

[SES]

There was a recent discussion on this site about constructing a phylogeny from SNPs that may be relevant. Hope that helps.