How to build rooted tree

[favite...@gmail.com]

I am wondering if building rooted trees (using/not using EPA algorithm) is included in raxml tutorial. I am able to make the unrooted trees but using the -o option doesn't give me a rooted tree. thanks

[Jacob Steenwyk]

Hi, 

There are three options I am aware of:

i)

You can root the tree using the -f I option

raxmlHPC ­-f I ­-m GTRCAT -­t unrootedTree -­n TEST

This will root the tree based on the branch that best balances subtree lengths.

ii)

You can root on a group or a single taxa using the -o option

Here, a tree is being rooted at Rat and Mouse so multiple taxa should be separated by a comma

raxmlHPC ­-s alg ­-p 12345 ­-m GTRGAMMA ­-o Rat,Mouse -­n TEST

iii)

Alternatively, you can root the tree on a specific taxa or clade in figtree.

best,

Jacob

On Mon, Apr 16, 2018 at 11:30 AM, <favite...@gmail.com> wrote:

I am wondering if building rooted trees (using/not using EPA algorithm) is included in raxml tutorial. I am able to make the unrooted trees but using the -o option doesn't give me a rooted tree. thanks

--
You received this message because you are subscribed to the Google Groups "raxml" group.
To unsubscribe from this group and stop receiving emails from it, send an email to raxml+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Alexandros Stamatakis]

Thanks Jacob,

Not much to add to this:

When using the EPA for placing outgroups:

1. Pick the branch with the highest placement probability of the outgroup

2. open a tree viewer

3. root the tree on that branch.

Be careful when re-rooting trees in tree viewers as the support values
might be flipped: https://academic.oup.com/mbe/article/34/6/1535/3077051

Also, tree rooting is always just a drawing option (except if you have a
non time-reversible model) the trees are always unrooted trees in the
mathematical sense.

Alexis

> send an email to raxml+un...@googlegroups.com
> <mailto:raxml+un...@googlegroups.com>.

> For more options, visit https://groups.google.com/d/optout

> <https://groups.google.com/d/optout>.

>
>
> --
> You received this message because you are subscribed to the Google
> Groups "raxml" group.
> To unsubscribe from this group and stop receiving emails from it, send

> an email to raxml+un...@googlegroups.com
> <mailto:raxml+un...@googlegroups.com>.

> For more options, visit https://groups.google.com/d/optout.

--
Alexandros (Alexis) Stamatakis

Research Group Leader, Heidelberg Institute for Theoretical Studies
Full Professor, Dept. of Informatics, Karlsruhe Institute of Technology

www.exelixis-lab.org

[Favites Test]

Thank you Jacob for your quick reply. I actually used the -f I option as Jacob suggested. It doesn't seem to be producing a rooted tree. When I fed the tree that RAxML labeled as rooted to pyhton package scikit-bio to compute UniFrac and it gives me error: 'Tree' object has no attribute 'root'

@Alexis: I have no clue how to use EPA. That's why I left this message here. Would be great if you could point me to the right resources. thanks

[Alexandros Stamatakis]

On 16.04.2018 21:26, Favites Test wrote:
> Thank you Jacob for your quick reply. I actually used the -f I option as
> Jacob suggested. It doesn't seem to be producing a rooted tree. When I
> fed the tree that RAxML labeled as rooted to pyhton package scikit-bio
> to compute UniFrac and it gives me error: 'Tree' object has no attribute
> 'root'

sounds maybe like a tree format issue, did you try looking at the tree
with dendroscope to see if it's rooted or not?

> @Alexis: I have noclue how to use EPA. That's why I left this message

> here. Would be great if you could point me to the right resources. thanks

Well, EPA is part of RAxML, so it's documented in the manual:

https://sco.h-its.org/exelixis/resource/download/NewManual.pdf

there are also quite some posts on the google group for which you can
search.

Finally, there's a all new efficient re-implementation of the EPA
available here:

https://github.com/Pbdas/epa-ng

Alexis

> it, send an email to raxml+un...@googlegroups.com
> <mailto:raxml+un...@googlegroups.com>.

> For more options, visit https://groups.google.com/d/optout

> <https://groups.google.com/d/optout>.

>
>
> --
> You received this message because you are subscribed to the Google
> Groups "raxml" group.
> To unsubscribe from this group and stop receiving emails from it,

> send an email to raxml+un...@googlegroups.com
> <mailto:raxml+un...@googlegroups.com>.

> For more options, visit https://groups.google.com/d/optout

> <https://groups.google.com/d/optout>.

>
>
> --
> You received this message because you are subscribed to the Google
> Groups "raxml" group.
> To unsubscribe from this group and stop receiving emails from it, send

> an email to raxml+un...@googlegroups.com
> <mailto:raxml+un...@googlegroups.com>.

> For more options, visit https://groups.google.com/d/optout.

--

[Favites Test]

I did look at the tree (with FigTree), It's not rooted. And I am wondering why... could it be because raxml couldn't find the branch lengths? Below is the command I used:

raxmlHPC -f I -m GTRCAT -t ./RAxML_bestTree$$$ -n  TEST

Will give the EPA a try. Searching though the posts here, was a bit confusing. Will have to get back to you if another problem comes up. Thank you!

        it, send an email to raxml+unsubscribe@googlegroups.com
        <mailto:raxml+unsubscribe@googlegroups.com>.

        For more options, visit https://groups.google.com/d/optout
        <https://groups.google.com/d/optout>.

    --     You received this message because you are subscribed to the Google
    Groups "raxml" group.
    To unsubscribe from this group and stop receiving emails from it,

    send an email to raxml+unsubscribe@googlegroups.com
    <mailto:raxml+unsubscribe@googlegroups.com>.

    For more options, visit https://groups.google.com/d/optout
    <https://groups.google.com/d/optout>.


--
You received this message because you are subscribed to the Google Groups "raxml" group.

To unsubscribe from this group and stop receiving emails from it, send an email to raxml+unsubscribe@googlegroups.com <mailto:raxml+unsubscribe@googlegroups.com>.

For more options, visit https://groups.google.com/d/optout.

--
Alexandros (Alexis) Stamatakis

Research Group Leader, Heidelberg Institute for Theoretical Studies
Full Professor, Dept. of Informatics, Karlsruhe Institute of Technology

www.exelixis-lab.org

--
You received this message because you are subscribed to the Google Groups "raxml" group.

To unsubscribe from this group and stop receiving emails from it, send an email to raxml+unsubscribe@googlegroups.com.

[alexandros.stamatakis]

That's weird. Can you send me the input tree such that I can check?

Alexis

Prof. Dr. Alexandros Stamatakis

www.exelixis-lab.org

-------- Ursprüngliche Nachricht --------

Von: Favites Test <favite...@gmail.com>

Datum: 16.04.18 21:49 (GMT+01:00)

An: ra...@googlegroups.com

Betreff: Re: [raxml] How to build rooted tree

To unsubscribe from this group and stop receiving emails from it, send an email to raxml+un...@googlegroups.com.

[Favites Test]

Absolutely! I will email you. thanks

On Mon, Apr 16, 2018 at 9:38 PM, alexandros.stamatakis <alexandros...@gmail.com> wrote:

That's weird. Can you send me the input tree such that I can check?

Alexis

Prof. Dr. Alexandros Stamatakis

www.exelixis-lab.org

-------- Ursprüngliche Nachricht --------

Von: Favites Test <favite...@gmail.com>

Datum: 16.04.18 21:49 (GMT+01:00)

An: ra...@googlegroups.com

Betreff: Re: [raxml] How to build rooted tree

I did look at the tree (with FigTree), It's not rooted. And I am wondering why... could it be because raxml couldn't find the branch lengths? Below is the command I used:

raxmlHPC -f I -m GTRCAT -t ./RAxML_bestTree$$$ -n  TEST

Will give the EPA a try. Searching though the posts here, was a bit confusing. Will have to get back to you if another problem comes up. Thank you!

[Alexis]

I just tested this with the file you sent me: 

 ./raxmlHPC-AVX -f I -m GTRCAT -t RAxML_bestTree.B_178.fa_seq -n TEST_R

the tree that is written to file: RAxML_rootedTree.TEST_R

is clearly rooted when viewed with Dendroscope (see attachment).

Alexis 

Am Dienstag, 17. April 2018 18:55:53 UTC+2 schrieb Favites Test:

Absolutely! I will email you. thanks

[Favites Test]

Thanks much for your quick response! I understand it looks like a rooted tree. Yet skbio.diversity package gives me an error such that: 'Tree' object has no attribute 'root'

I'm following up with them. If there is sth wrong with the way I read the tree or a mistake like that, I'll clarify it on the google group. Thanks again!