7
$\begingroup$

I understand that rooted and unrooted phylogenetic trees are similar in structure

But how can they be easily identified as one or the other?

Is it simply based on the presence or lack of a named root (the root is identified as a particular ancestor) or outgroup?

$\endgroup$
2
  • $\begingroup$ At which "stage" do you want to identify the tree? When it's a file (Newick), or when it's a drawn figure? $\endgroup$ Dec 15, 2011 at 8:32
  • $\begingroup$ I'd say a drawn figure? Like something you would find on a google search for phylogenetic tree (images.google.com/search?tbm=isch&q=phylogenetic%20tree) $\endgroup$
    – arcyqwerty
    Dec 15, 2011 at 8:36

2 Answers 2

6
$\begingroup$

In case of a drawn figure, either look for the a short stem to the left, or for an explicitly labelled outgroup. Trees derived from phylogenetic analyses are normally unrooted by default, you need to root them by either making an arbitrary (but hopefully informed) decision or by adding an outgroup.

$\endgroup$
5
  • $\begingroup$ So when you say root it by making an arbitrary decision, does that mean labeling a particular ancestral individual as a root? Also, does this consider midpoint rooting? $\endgroup$
    – arcyqwerty
    Dec 15, 2011 at 8:43
  • $\begingroup$ Midpoint root is certainly one of the ways you can choose to root your tree. IMHO rooting a tree of species or proteins without an outgroup is done by adding an extra root node. $\endgroup$ Dec 15, 2011 at 8:45
  • $\begingroup$ And just to clarify, how would the 'extra root node' appear on a diagram? As just a labeled point from where the rest of the tree branches off from? $\endgroup$
    – arcyqwerty
    Dec 15, 2011 at 8:47
  • $\begingroup$ Usually, there's a short stem going in a different direction than all the other leaves. But a labelled point is also fine, I guess $\endgroup$ Dec 15, 2011 at 8:51
  • $\begingroup$ Unrooted by default? Which algorithms are you talking about? $\endgroup$
    – user120
    Dec 15, 2011 at 9:00
2
$\begingroup$

Trees can not be distinguished from each other because every unrooted tree can be made rooted by adding a root and vice versa by deletion.

If you mean dendrograms... the root is always the one node where the distance to each leaf is the same. But I don't know of an unrooted dendrogram.

$\endgroup$
2
  • $\begingroup$ Yes, but how would I be able to tell whether or not a tree has a root Like how to tell between an unrooted tree and a midpoint rooted tree? $\endgroup$
    – arcyqwerty
    Dec 15, 2011 at 8:30
  • $\begingroup$ If you really mean trees. They are not distinguishable. $\endgroup$
    – user120
    Dec 15, 2011 at 8:36

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .