Transcript or cds for phylogenetic tree ?

Question

[Picasa]

Hi,

I would like to make a single phylogenetic tree after a multiple alignment of sequences of differents species.

For the sequences, should I use the whole transcript or only de CDS part ? or it doesn't matter.

Thanks for your help.

Answer 1

[lieven.sterck]

It does matter.

Go for the CDS (or the protein is even better). Evolutionary constraints/conservation works mainly on those sequences (and less on the UTR, which is what you add when using the complete mRNA/transcripts).

If in the unlikely event that the CDS/protein does not give you enough 'resolution' you might consider adding UTRs to the analysis but keep in mind that comes with a whole bunch of other issues :/ (you might even go very wild and add all the introns, but let's no go there yet :) )

[Picasa]

Thanks for your answer.

Evolutionary constraints/conservation works mainly on those sequences

I am not sure to understand this. Do you mean algorithm used by phylogenetic softwares (such as raxml or phyml or whatever) used model that worked with cds (dna level) and/or amino acid ?

[lieven.sterck]

well, yes and no.

I meant that the biological constraints put into place by evolution mainly play on those kind of sequences. As a consequence the algorithms that model/analyse those things are also best performing (optimised) on those sequences indeed.