How to make a rooted tree ultrametric

667 views
Skip to first unread message

Ben Fulton

unread,
May 5, 2017, 4:57:53 PM5/5/17
to hahnlab-cafe

Here's another Biostars question from the same user (who gave me a nice upvote for my answer btw, so thanks for helping):


I have used OrthoFinder to identify orthogroups among five species. Part of the output is a rooted phylogenetic tree, but I think it is not ultrametric?


I would like to use the tree as part of the input for a CAFE analysis. The CAFE tutorial described using the program r8s to make a tree ultrametric, but I don't know the number of sites in the alignment used to estimate the species tree. TimeTree has approximate divergence time for some of the species in my tree, which I can use to calibrate the tree?


Any suggestions on how I can create an ultrametric phylogenetic tree would be greatly appreciated!


https://www.biostars.org/p/251282/


Thanks!


--

Ben Fulton

Research Technologies

Scientific Applications and Performance Tuning

Indiana University

Hahn, Matthew

unread,
May 5, 2017, 5:32:25 PM5/5/17
to Ben Fulton, hahnlab-cafe
Fabio or Gregg would be better to answer this....
--
You received this message because you are subscribed to the Google Groups "hahnlab-cafe" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hahnlabcafe...@googlegroups.com.
To post to this group, send email to hahnl...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/hahnlabcafe/31974514-86ff-4612-a7a7-895675f27c61%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Gregg Thomas

unread,
May 6, 2017, 3:00:20 PM5/6/17
to Hahn, Matthew, Ben Fulton, hahnlab-cafe
I'm not familiar with OrthoFinder, but it appears to just be a wrapper for a bunch of other things, using BLAST and MCL for the clustering. It then also aligns (MAFFT), builds gene trees (FastTree). It does NOT say how it infers the species tree.

That being said, I have to assume that somewhere in the output files are the alignments for the gene trees, or the concatenated alignment used to make the species tree (again, not sure this is what it does). It should be easy enough to count the number of sites in those alignments. Then she would have to pick at least one calibration point, possibly from TimeTree, for r8s.

-Gregg

On Fri, May 5, 2017 at 5:32 PM, Hahn, Matthew <m...@indiana.edu> wrote:
Fabio or Gregg would be better to answer this....

On May 5, 2017, at 4:57 PM, Ben Fulton <benmar...@gmail.com> wrote:

Here's another Biostars question from the same user (who gave me a nice upvote for my answer btw, so thanks for helping):


I have used OrthoFinder to identify orthogroups among five species. Part of the output is a rooted phylogenetic tree, but I think it is not ultrametric?


I would like to use the tree as part of the input for a CAFE analysis. The CAFE tutorial described using the program r8s to make a tree ultrametric, but I don't know the number of sites in the alignment used to estimate the species tree. TimeTree has approximate divergence time for some of the species in my tree, which I can use to calibrate the tree?


Any suggestions on how I can create an ultrametric phylogenetic tree would be greatly appreciated!


https://www.biostars.org/p/251282/


Thanks!


--

Ben Fulton

Research Technologies

Scientific Applications and Performance Tuning

Indiana University

--
You received this message because you are subscribed to the Google Groups "hahnlab-cafe" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hahnlabcafe+unsubscribe@googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "hahnlab-cafe" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hahnlabcafe+unsubscribe@googlegroups.com.

To post to this group, send email to hahnl...@googlegroups.com.

Gregg Thomas

unread,
May 6, 2017, 3:02:59 PM5/6/17
to Hahn, Matthew, Ben Fulton, hahnlab-cafe
There is an issue in the OrthoFinder github posted 10 days ago asking how the species tree is inferred. It is still unanswered:

https://github.com/davidemms/OrthoFinder/issues/62

Nancy Dong

unread,
May 7, 2017, 5:49:07 PM5/7/17
to hahnlab-cafe
Hello!

I am the original poster on Biostars. I think this question can be closed, as I used MEGA7 to make an ultrametric tree using single copy orthologs identified by OrthoFinder.

The new problem is now that the tree generated by MEGA7 is not "accepted" by CAFE. I will make a new post for that question.

Thank you very much everyone!

Nancy
Reply all
Reply to author
Forward
0 new messages