[Jalview-discuss] about tree

[Jim Procter]

On 24/06/2011 18:57, Chen, Xianfeng (CDC/OID/NCIRD) (CTR) wrote:
> Thanks for your kindly response.
you're very welcome !
> Jalview is perfect in presenting alignment, but I want to sub-group sequences in multialignment based on their similarity, in the other words, I will take look at the phylogenetic tree and decide what sequences to be selected, therefore I need to know some parameters building the tree. Dendroscope and others are good tree viewer, but could allow you to sub-group sequence and re-multiple-alignment.
Ah - I see, so Jalview is certainly what you want. However, it sounds 
like you should calculate the tree using another program and import it 
onto the source alignment in Jalview (which is probably what you are 
doing already). You can then use Jalview's tree-based subgrouping 
features to explore the variation and conservation patterns between the 
subgroups (there's an exercise in chapter 2 of the jalview manual at 
http://www.jalview.org/tutorial/TheJalviewTutorial_screen.pdf that 
covers this).

Co-incidentally, I have been in a few discussions with people here in 
Dundee about introducing 'real' phylogenetic methods into Jalview, and 
hope that we'll have at least one phylogenetic method available in a 
release by the end of the year. Until then, you'll have to use another 
program to calculate a reliable tree. For this, I'd normally prefer to 
use TOPAli or SplitsTree - which both allow trees to be saved in newick 
format for import to Jalview.

Happy jalviewing!
Jim.