rerooting trifurcated tree issue about dendropy

[jeetsukumaran]

I will look into this in detail tomorrow, but at first glance I rather
suspect that the problem might be that you are reading in the tree as
unrooted. Try adding a "[&R]" to your tree string and see if it works?
It still might a bug on the library-side, even if it works with the
above work-around, depending on what might be reasonably expected in
this case ...

On 9/8/16 9:35 PM, geronimp wrote:

Hi there,

I'm encountering a bit of a problem where I'm trying to reroot a tree
that is trifurcated at the root so that the new root is on the longest
edge descending from the current root. I've outlined the problem with
the following code:

example tree

|import dendropy tree_string =
'(1111319:0.0267,(1111717:0.01494,1111294:0.02598)1.000:0.02879,(1111874:0.03322,1111783:0.01707)0.772:0.01236);'
tree = dendropy.Tree.get(data=tree_string, schema="newick") |

The tree is trifurcated

|print tree.as_ascii_plot(plot_metric='length')
/------------------------------ 1111319 | | /----------------- 1111717
+--------------------------------+ | ------------------------------
1111294 | | /-------------------------------------- 1111874
-------------+ ------------------- 1111783 |

I want to reroot the tree on the longest edge descending from the root
(i.e. the middle one). So I find that edge, and attempt to reroot it on
that edge

|children = {x.length:x for x in tree.seed_node.child_edges()}
longest_edge_length = max(children.keys()) longest_edge =
children[longest_edge_length] tree.reroot_at_edge(longest_edge,
length1=(longest_edge_length/2), length2=(longest_edge_length/2)) |

But the tree is still trifurcated

|print tree.as_ascii_plot(plot_metric='length') /-----------------
1111717 /--------------------------------+ |
------------------------------ 1111294 |
+------------------------------ 1111319 | |
/-------------------------------------- 1111874 -------------+
------------------- 1111783 |

Strangely enough, when I try it again on the just-rerooted tree it works

|print tree.as_ascii_plot(plot_metric='length')
tree.reroot_at_edge(longest_edge, length1=(longest_edge_length/2),
length2=(longest_edge_length/2)) /--------------- 1111717
/--------------+ | --------------------------- 1111294 + |
/--------------------------- 1111319 --------------+ |
/----------------------------------- 1111874 ------------+
----------------- 1111783 |

The following description of the reroot_at_edge function lead me to
believe that it would create a node on the edge you specify, and reroot
the tree on that node, with the distances on either side of the node
being specified with the 'length' parameters:

Takes an internal edge, |edge|, adds a new node to it, and then roots
the tree on the new node.

I'm not totally sure if this is a bug or just my misunderstanding of how
to use the reroot_at_edge function. Do you have any idea why it would
not be working on the first try? How would I be able to achieve what I
need on the first go?

I'm using version 4.1.0 of dendropy

Thanks so much for any help you can provide.

Joel

—
You are receiving this because you are subscribed to this thread.
Reply to this email directly, view it on GitHub
#69, or mute the
thread
https://github.com/notifications/unsubscribe-auth/AABmR4ixhdBLbGcZpJeu4CsrUZF1QNzRks5qoLfWgaJpZM4J4nX5.

---

Jeet Sukumaran

[email protected]

Blog/Personal Pages:
http://jeetworks.org/
GitHub Repositories:
http://github.com/jeetsukumaran
Photographs (as stream):
http://www.flickr.com/photos/jeetsukumaran/
Photographs (by galleries):

http://www.flickr.com/photos/jeetsukumaran/sets/

[geronimp]

You're right, seems I wasn't checking whether the tree was rooted.
Thanks for the quick reply!