Statistical support for a clade with the exact internal topology

14 views
Skip to first unread message

Alexei Kassian

unread,
Mar 29, 2019, 8:40:37 AM3/29/19
to DendroPy Users
Hi all,

I have a set of trees (e.g., MrBayes output *.runX.t) with many taxa: A, B, C, .... Z.

I need to infer statistical support for a certain clade ((E,F),G).

I use the options '--target-tree-filepath=target.tre --force-unrooted' and the target.tre with the following content: (A,B,C,...Z,((E,F),G)).

For the clade ((E,F),G), Sumtrees returns the cumulative number of trees which contain ((E,F),G), or (E,(F,G)), or (E,F,G). 

Is it possible to get a number of trees with the exact clade ((E,F),G) without trees with (E,(F,G)) or (E,F,G)?

Thank you in advance.

Jeet Sukumaran

unread,
Mar 29, 2019, 9:12:14 PM3/29/19
to dendrop...@googlegroups.com, Alexei Kassian
Would '--force-rooted' give you what you want?
> --
> You received this message because you are subscribed to the Google
> Groups "DendroPy Users" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to dendropy-user...@googlegroups.com
> <mailto:dendropy-user...@googlegroups.com>.
> For more options, visit https://groups.google.com/d/optout.

--

----------------------------------------------------
Jeet Sukumaran
----------------------------------------------------
Assistant Professor
Biology Department
San Diego State University
----------------------------------------------------
Lab:
https://sukumaranlab.org/
Blog:
https://jeetblogs.org/
Repositories:
https://github.com/jeetsukumaran
Photography:
https://www.flickr.com/photos/jeetsukumaran/
Calendar:
https://goo.gl/dG5Axs
----------------------------------------------------
Email:
jsuku...@sdsu.edu (work)
jeetsu...@gmail.com (personal)
----------------------------------------------------
Mailing Address:
Biology Department, LS 262
San Diego State University
5500 Campanile Drive
San Diego, CA 92182-4614
----------------------------------------------------

Alexei Kassian

unread,
Mar 30, 2019, 5:38:25 AM3/30/19
to DendroPy Users
Unfortunately not. The key '--force-rooted' returns the same cumulative number.


суббота, 30 марта 2019 г., 4:12:14 UTC+3 пользователь Jeet Sukumaran написал:
Would '--force-rooted' give you what you want?

On 3/29/19 5:40 AM, Alexei Kassian wrote:
> Hi all,
>
> I have a set of trees (e.g., MrBayes output *.runX.t) with many taxa: A,
> B, C, .... Z.
>
> I need to infer statistical support for a certain clade ((E,F),G).
>
> I use the options '--target-tree-filepath=target.tre --force-unrooted'
> and the target.tre with the following content: (A,B,C,...Z,((E,F),G)).
>
> For the clade ((E,F),G), Sumtrees returns the cumulative number of trees
> which contain ((E,F),G), or (E,(F,G)), or (E,F,G).
>
> Is it possible to get a number of trees with the exact clade ((E,F),G)
> without trees with (E,(F,G)) or (E,F,G)?
>
> Thank you in advance.
>
> --
> You received this message because you are subscribed to the Google
> Groups "DendroPy Users" group.
> To unsubscribe from this group and stop receiving emails from it, send

Jeet Sukumaran

unread,
Mar 30, 2019, 11:44:44 AM3/30/19
to dendrop...@googlegroups.com, Alexei Kassian
Could you filter out the topologies with "(E,F,G)" and "(E,(F,G))"?

-- jeet
> > an email to dendrop...@googlegroups.com <javascript:>
> > <mailto:dendropy-user...@googlegroups.com <javascript:>>.
> > For more options, visit https://groups.google.com/d/optout
> <https://groups.google.com/d/optout>.
>
> --
>
> ----------------------------------------------------
> Jeet Sukumaran
> ----------------------------------------------------
> Assistant Professor
> Biology Department
> San Diego State University
> ----------------------------------------------------
> Lab:
> https://sukumaranlab.org/
> Blog:
> https://jeetblogs.org/
> Repositories:
> https://github.com/jeetsukumaran <https://github.com/jeetsukumaran>
> Photography:
> https://www.flickr.com/photos/jeetsukumaran/
> <https://www.flickr.com/photos/jeetsukumaran/>
> Calendar:
> https://goo.gl/dG5Axs
> ----------------------------------------------------
> Email:
> jsuku...@sdsu.edu <javascript:> (work)
> jeetsu...@gmail.com <javascript:> (personal)
> ----------------------------------------------------
> Mailing Address:
>      Biology Department, LS 262
>      San Diego State University
>      5500 Campanile Drive
>      San Diego, CA 92182-4614
> ----------------------------------------------------
>
> --
> You received this message because you are subscribed to the Google
> Groups "DendroPy Users" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to dendropy-user...@googlegroups.com
> <mailto:dendropy-user...@googlegroups.com>.

Alexei Kassian

unread,
Mar 30, 2019, 2:42:20 PM3/30/19
to DendroPy Users
How can I do so?
E.g., I use target.tre with (A,B,C,...Z,((E,F),G)).
Sumtrees reports 0.7 for the clade E+F+G (whatever internal structure it has) and 0.95 for (E,F). But it is not necessarily that every tree with (E,F) has the clade ((E,F),G).



суббота, 30 марта 2019 г., 18:44:44 UTC+3 пользователь Jeet Sukumaran написал:
>      > <mailto:dendropy-users+unsub...@googlegroups.com <javascript:>>.
Message has been deleted

Alexei Kassian

unread,
Mar 31, 2019, 11:27:43 AM3/31/19
to DendroPy Users

I can reformulate the problem, since my actual task is somewhat more complicated than the simplest task described in my first message.

I have a poorly resolved 'roaming' taxon Z which skips across the tree, particularly Z can be inserted within the clade E,F,G.

I want to infer probability of my clade ((E,F),G) regardless whether the taxon Z is placed within this clade or not.

To do so, I need to infer:

1)   ((E,F),G)

2)   ((E,F,Z),G)

3)   ((E,F),G,Z)

and some other combinations with Z. And then summarize the obtained values.

Does Sumtrees allow such calculations?



суббота, 30 марта 2019 г., 21:42:20 UTC+3 пользователь Alexei Kassian написал:

Jeet Sukumaran

unread,
Mar 31, 2019, 10:48:43 PM3/31/19
to dendrop...@googlegroups.com, Alexei Kassian
You can do this by rolling your own summarization script using the same
building blocks from DendroPy that SumTrees does.

Short of that, perhaps a workaround would be to run sumtrees and ask it
to output extended information using the "-x extra" flag. Then look into
the "extra.bipartitions.tsv" "extra.bipartitions.trees" files that
result and sum up the probabilities of the clades that you are
interested in that marginalize over Z's position/presence.

-- jeet
> >      > <mailto:dendropy-user...@googlegroups.com
> > <mailto:dendropy-user...@googlegroups.com>.
> > For more options, visit https://groups.google.com/d/optout
> <https://groups.google.com/d/optout>.
>
> --
>
> ----------------------------------------------------
> Jeet Sukumaran
> ----------------------------------------------------
> Assistant Professor
> Biology Department
> San Diego State University
> ----------------------------------------------------
> Lab:
> https://sukumaranlab.org/
> Blog:
> https://jeetblogs.org/
> Repositories:
> https://github.com/jeetsukumaran <https://github.com/jeetsukumaran>
> Photography:
> https://www.flickr.com/photos/jeetsukumaran/
> <https://www.flickr.com/photos/jeetsukumaran/>
> Calendar:
> https://goo.gl/dG5Axs
> ----------------------------------------------------
> Email:
> jsuku...@sdsu.edu (work)
> jeetsu...@gmail.com (personal)
> ----------------------------------------------------
> Mailing Address:
>      Biology Department, LS 262
>      San Diego State University
>      5500 Campanile Drive
>      San Diego, CA 92182-4614
> ----------------------------------------------------
>
> --
> You received this message because you are subscribed to the Google
> Groups "DendroPy Users" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to dendropy-user...@googlegroups.com
> <mailto:dendropy-user...@googlegroups.com>.

Alexei Kassian

unread,
Apr 1, 2019, 6:57:13 PM4/1/19
to DendroPy Users
Dear Jeet,
thank you. I'll try to follow this way. I also think such a new function, be it added in Sumtrees, would be useful for many Denropy users.

понедельник, 1 апреля 2019 г., 5:48:43 UTC+3 пользователь Jeet Sukumaran написал:
>          >      > <mailto:dendropy-users+unsub...@googlegroups.com
>          > <mailto:dendropy-users+unsub...@googlegroups.com>.
Reply all
Reply to author
Forward
0 new messages