How to root this tree?
1
0
Entering edit mode
6 weeks ago
Mirko • 0

I made a phylogenetic tree using RAxML using ML estimation method and it looks like this
Original unrooted tree The visualization is from iTOL using the unrooted display option. The tree is unrooted (i guess, i never set any outgroup) and i need to root it.
All the species are macro organism except for 2 that are Platyhelminthes. You can recognize those 2 species because they are very far from the other species.
I know the basic knowledge behind the rooting of a tree but i am not sure whether what i want to do is biologically sound or doable and how to do it. My idea so far is that i want to transform the tree like i sketched in this image
sketch
So basically the outgroup is not a single species but the clade of taxa T1 and T2 (clade called C1).

  1. Can i do it like this? Is it biologically reasonable
  2. How to do it? I mainly use Biopython and RAxML but i can use any other software to do it.
Phylogeny Tree Root • 442 views
ADD COMMENT
1
Entering edit mode
6 weeks ago
Michael 54k

I think you can root the tree using the Platyhelminthes as an out-group. You can also try out different options and see if the main grouping is unaffected. It is common to use multiple taxa as out-groups. In FigTree, you can simply do this by clicking the "set root" button and then the branch, in the R package ape there is a root function: Rooting phylogenetic trees in R using Ape's root function: Trees don't seem to root. I don't know if ITOL can put a root on a clade or only on a single taxon. But even if not, you can root your tree outside and open it in ITOL again.

Whether rooting makes sense on whether your out-group still contains "true" orthologues (and if the divergence marks the earliest divergence in the lineage evolution), but if they were not true orthologues, they should not have been included in the first place.

ADD COMMENT
0
Entering edit mode

Thanks for the clarification. I actually missed to mention one of the main problems: my first split is a trifurcating node. I don't know why but the first node has a branch length of 0 and thus the first split is a trifurcation.

this is what the start of the tree looks like (the green line is drawn by me it's not actually in the tree) enter image description here

Can i set the root to the midpoint of the (Echinostoma, Schistostoma) clade and the rest of the tree (the green line i drew on the image)? Maybe you already answered this but i just want to make sure i am interpreting the tree right.

ADD REPLY
0
Entering edit mode

This could indicate that something is wrong with your alignment. Is this based on protein alignments or are these probably not orthologues at all?

ADD REPLY
0
Entering edit mode

Alignment is done on the COX1 genes (DNA) extracted from a curated dataset using the same region. Also the rest of the tree looks very good. I was wondering if this is actually something that should always happen when making an unrooted tree. At first i was fooled by the rectangular representation of my unrooted tree, but if you maintain the topology and draw it in a more standard unrooted style it looks exactly like the unrooted tree in my sketch. The green line would then be added between the C1 and C2 nodes.

Also in Biopython when i use the midpoint rooting procedure i get exactly the rooted tree i was looking for.

ADD REPLY
1
Entering edit mode

I think you should not try to build a DNA-based phylogeny of distantly related species. You should try using either a codon model or the AA sequences instead. I am assuming that your tree would be much more like a tree then. I am guessing that the tree looks like that because there is very low sequence similarity on the DNA level between these clades, such that already the alignment doesn't work. You need to either shift to AA sequences or choose a more closely related outgroup.

ADD REPLY

Login before adding your answer.

Traffic: 1623 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6