Re: [R-sig-phylo] Using results of make.simmap to color edges by most likely state

[Rafael S Marcondes]

Thanks, Liam! It worked perfectly.

-Rafa

*--*
*Rafael S. Marcondes, Ph.D.*
*https://www.rafaelmarcondes.com/ *
Faculty Fellow in EEB
Department of BioSciences
Rice University
Houston TX 77005

Pronouns: he/him


*"Eu quase que nada não sei. Mas desconfio de muita coisa"*
*"I almost don't know nothing. But I suspect many things"*
  -João Guimarães Rosa, Brazilian novelist
(Portuguese original and free English translation by me)


On Tue, Apr 18, 2023 at 1:35 PM Liam J. Revell  wrote:

> Dear Rafa & list.
>
> Here is a solution:
> http://blog.phytools.org/2023/04/coloring-edges-of-plotted-tree-by.html.
>
> I applied it to marginal ancestral states obtained using the new
> *phytools* method *ancr*, but the same principle could be used with a
> summary of stochastic maps. I will make a subsequent follow-up post
> covering that.
>
> Let us know if this works!
>
> All the best, Liam
> Liam J. Revell
> Professor of Biology, University of Massachusetts Boston
> Web: http://faculty.umb.edu/liam.revell/
> Book: Phylogenetic Comparative Methods in R
> 
> (*Princeton University Press*, 2022)
>
>
> On 4/18/2023 1:21 PM, Liam J. Revell wrote:
>
> Dear Rafa.
>
> Just a quick comment re the comment "I also noticed that results of simmap
> have one extra edge than the tree
> itself"
>
> From ancestral state reconstruction (marginal reconstruction or a summary
> of stochastic mapping) you'll end up with a matrix containing a number of
> rows equal to the number of internal nodes of the tree, or (in your case)
> the number of tips + the number of internal nodes. In your case, you're
> interested in coloring all the edges of the tree, which will be one fewer
> than the number of nodes + tips. This is because every node (or tip) is
> preceded by one edge: except for the root. Likewise, the order of your
> ancestral state matrix is by node index; whereas the edges of the tree can
> have a variety of different orders.
>
> Let me post a solution to my blog quickly & then I'll share it with this
> group.
>
> All the best, Liam
> Liam J. Revell
> Professor of Biology, University of Massachusetts Boston
> Web: http://faculty.umb.edu/liam.revell/
> Book: Phylogenetic Comparative Methods in R
> 
> (*Princeton University Press*, 2022)
>
>
> On 4/18/2023 10:59 AM, Rafael S Marcondes wrote:
>
> CAUTION: EXTERNAL SENDER
>
> Hi all,
>
> I would like to use the results of make.simmap to plot the edges of a tree
> by their most likely state. (Because my tree is really large and plotting
> pie charts at the nodes gets really polluted.)
>
> So first I took $ace from the results to describe.simmap and added a column
> for the most likely state, and another column for the color for plotting:
>
>
> head(ace.max.lik)
>
>  CLEL M P SL maxlikcol
> 6621  0 0.000 0.002 0.998  0  P darkgoldenrod3
> 6622  0 0.000 0.001 0.999  0  P darkgoldenrod3
> 6623  0 0.000 0.007 0.993  0  P darkgoldenrod3
> 6624  0 0.000 0.004 0.996  0  P darkgoldenrod3
> 6625  0 0.000 0.001 0.999  0  P darkgoldenrod3
> 6626  0 0.001 0.020 0.979  0  P darkgoldenrod3
>
> And I separately did that for the tips. (Which I also estimated, since they
> had uncertainty in the original data):
>
>  CL EL M P SL maxlikcol
> Rhea_americana  0.0000.000   0.000   1.000
>0.000  P darkgoldenrod3
> Rhea_pennata0.0000.000   0.008   0.992
>0.000  P darkgoldenrod3
> Casuarius_casuarius 0.0000.004   0.074   0.922
>0.000  P darkgoldenrod3
> Casuarius_bennetti  0.0000.010   0.132   0.858
>0.000  P darkgoldenrod3
> Dromaius_novaehollandiae0.0010.000   0.045   0.953
>0.001  P darkgoldenrod3
> Apteryx_owenii  0.0000.000   0.995   0.005
>0.000  M   blue
>
> Then I tried to plot the edges by color using ape, putting the tips in
> front of the internal nodes because that's the typical order of tree$edges:
>
> plot.phylo(x=m.ard[[1]],cex=0.2,edge.width=0.01,type='fan',
>edge.lty=3,edge.color=c(tip.max.lik$col, ace.max.lik$col))
>
> It plots a beautiful plot, by the colors are mismatched to the edges. I
> also noticed that results of simmap have one extra edge than the tree
> itself:
>
>
> nrow(md.ard$ace)
>
> [1] 13239
>
>
> nrow(m.ard[[1]]$edge)
>
> [1] 13238
>
> I suspect that's that one extra edge that's causing the mismatch, but I
> can't figure out where it's coming from, or how to fix it. Any tips or
> workarounds to accomplish this task, which I might be overcomplicating?
>
> PS: I'm well aware that it's a bad idea to "flatten" the likelihoods of
> different states 

Re: [R-sig-phylo] Using results of make.simmap to color edges by most likely state

[Liam J. Revell]

Dear Rafa & list.

Here is a solution: 
http://blog.phytools.org/2023/04/coloring-edges-of-plotted-tree-by.html.

I applied it to marginal ancestral states obtained using the new 
/phytools/ method /ancr/, but the same principle could be used with a 
summary of stochastic maps. I will make a subsequent follow-up post 
covering that.

Let us know if this works!

All the best, Liam

Liam J. Revell
Professor of Biology, University of Massachusetts Boston
Web: http://faculty.umb.edu/liam.revell/
Book: Phylogenetic Comparative Methods in R 
 
(/Princeton University Press/, 2022)


On 4/18/2023 1:21 PM, Liam J. Revell wrote:
>
> Dear Rafa.
>
> Just a quick comment re the comment "I also noticed that results of 
> simmap have one extra edge than the tree
> itself"
>
> From ancestral state reconstruction (marginal reconstruction or a 
> summary of stochastic mapping) you'll end up with a matrix containing 
> a number of rows equal to the number of internal nodes of the tree, or 
> (in your case) the number of tips + the number of internal nodes. In 
> your case, you're interested in coloring all the edges of the tree, 
> which will be one fewer than the number of nodes + tips. This is 
> because every node (or tip) is preceded by one edge: except for the 
> root. Likewise, the order of your ancestral state matrix is by node 
> index; whereas the edges of the tree can have a variety of different 
> orders.
>
> Let me post a solution to my blog quickly & then I'll share it with 
> this group.
>
> All the best, Liam
>
> Liam J. Revell
> Professor of Biology, University of Massachusetts Boston
> Web: http://faculty.umb.edu/liam.revell/
> Book: Phylogenetic Comparative Methods in R 
>  
> (/Princeton University Press/, 2022)
>
>
> On 4/18/2023 10:59 AM, Rafael S Marcondes wrote:
>> CAUTION: EXTERNAL SENDER
>>
>> Hi all,
>>
>> I would like to use the results of make.simmap to plot the edges of a tree
>> by their most likely state. (Because my tree is really large and plotting
>> pie charts at the nodes gets really polluted.)
>>
>> So first I took $ace from the results to describe.simmap and added a column
>> for the most likely state, and another column for the color for plotting:
>>
>>> head(ace.max.lik)
>>   CLEL M P SL maxlikcol
>> 6621  0 0.000 0.002 0.998  0  P darkgoldenrod3
>> 6622  0 0.000 0.001 0.999  0  P darkgoldenrod3
>> 6623  0 0.000 0.007 0.993  0  P darkgoldenrod3
>> 6624  0 0.000 0.004 0.996  0  P darkgoldenrod3
>> 6625  0 0.000 0.001 0.999  0  P darkgoldenrod3
>> 6626  0 0.001 0.020 0.979  0  P darkgoldenrod3
>>
>> And I separately did that for the tips. (Which I also estimated, since they
>> had uncertainty in the original data):
>>
>>   CL EL M P SL maxlikcol
>> Rhea_americana  0.0000.000   0.000   1.000
>> 0.000  P darkgoldenrod3
>> Rhea_pennata0.0000.000   0.008   0.992
>> 0.000  P darkgoldenrod3
>> Casuarius_casuarius 0.0000.004   0.074   0.922
>> 0.000  P darkgoldenrod3
>> Casuarius_bennetti  0.0000.010   0.132   0.858
>> 0.000  P darkgoldenrod3
>> Dromaius_novaehollandiae0.0010.000   0.045   0.953
>> 0.001  P darkgoldenrod3
>> Apteryx_owenii  0.0000.000   0.995   0.005
>> 0.000  M   blue
>>
>> Then I tried to plot the edges by color using ape, putting the tips in
>> front of the internal nodes because that's the typical order of tree$edges:
>>
>> plot.phylo(x=m.ard[[1]],cex=0.2,edge.width=0.01,type='fan',
>> edge.lty=3,edge.color=c(tip.max.lik$col, ace.max.lik$col))
>>
>> It plots a beautiful plot, by the colors are mismatched to the edges. I
>> also noticed that results of simmap have one extra edge than the tree
>> itself:
>>
>>> nrow(md.ard$ace)
>> [1] 13239
>>
>>> nrow(m.ard[[1]]$edge)
>> [1] 13238
>>
>> I suspect that's that one extra edge that's causing the mismatch, but I
>> can't figure out where it's coming from, or how to fix it. Any tips or
>> workarounds to accomplish this task, which I might be overcomplicating?
>>
>> PS: I'm well aware that it's a bad idea to "flatten" the likelihoods of
>> different states into a single most likely state, and I absolutely do not
>> plan to discard that uncertainty. I just want to plot a simpler tree for
>> visualization.
>>
>> Thanks!
>>
>> -Rafa
>>
>> *--*
>> *Rafael S. Marcondes, Ph.D.*
>> *https://www.rafaelmarcondes.com/  *
>> Faculty Fellow in EEB
>> Department of BioSciences
>> Rice University
>> Houston TX 77005
>>
>> Pronouns: he/him
>>
>>
>> *"Eu quase que nada não sei. Mas desconfio de muita coisa"*
>> 

[R-sig-phylo] Using results of make.simmap to color edges by most likely state

[Rafael S Marcondes]

Hi all,

I would like to use the results of make.simmap to plot the edges of a tree
by their most likely state. (Because my tree is really large and plotting
pie charts at the nodes gets really polluted.)

So first I took $ace from the results to describe.simmap and added a column
for the most likely state, and another column for the color for plotting:

> head(ace.max.lik)
 CLEL M P SL maxlikcol
6621  0 0.000 0.002 0.998  0  P darkgoldenrod3
6622  0 0.000 0.001 0.999  0  P darkgoldenrod3
6623  0 0.000 0.007 0.993  0  P darkgoldenrod3
6624  0 0.000 0.004 0.996  0  P darkgoldenrod3
6625  0 0.000 0.001 0.999  0  P darkgoldenrod3
6626  0 0.001 0.020 0.979  0  P darkgoldenrod3

And I separately did that for the tips. (Which I also estimated, since they
had uncertainty in the original data):

 CL EL M P SL maxlikcol
Rhea_americana  0.0000.000   0.000   1.000
   0.000  P darkgoldenrod3
Rhea_pennata0.0000.000   0.008   0.992
   0.000  P darkgoldenrod3
Casuarius_casuarius 0.0000.004   0.074   0.922
   0.000  P darkgoldenrod3
Casuarius_bennetti  0.0000.010   0.132   0.858
   0.000  P darkgoldenrod3
Dromaius_novaehollandiae0.0010.000   0.045   0.953
   0.001  P darkgoldenrod3
Apteryx_owenii  0.0000.000   0.995   0.005
   0.000  M   blue

Then I tried to plot the edges by color using ape, putting the tips in
front of the internal nodes because that's the typical order of tree$edges:

plot.phylo(x=m.ard[[1]],cex=0.2,edge.width=0.01,type='fan',
   edge.lty=3,edge.color=c(tip.max.lik$col, ace.max.lik$col))

It plots a beautiful plot, by the colors are mismatched to the edges. I
also noticed that results of simmap have one extra edge than the tree
itself:

> nrow(md.ard$ace)
[1] 13239

> nrow(m.ard[[1]]$edge)
[1] 13238

I suspect that's that one extra edge that's causing the mismatch, but I
can't figure out where it's coming from, or how to fix it. Any tips or
workarounds to accomplish this task, which I might be overcomplicating?

PS: I'm well aware that it's a bad idea to "flatten" the likelihoods of
different states into a single most likely state, and I absolutely do not
plan to discard that uncertainty. I just want to plot a simpler tree for
visualization.

Thanks!

-Rafa

*--*
*Rafael S. Marcondes, Ph.D.*
*https://www.rafaelmarcondes.com/ *
Faculty Fellow in EEB
Department of BioSciences
Rice University
Houston TX 77005

Pronouns: he/him


*"Eu quase que nada não sei. Mas desconfio de muita coisa"*
*"I almost don't know nothing. But I suspect many things"*
  -João Guimarães Rosa, Brazilian novelist
(Portuguese original and free English translation by me)

[[alternative HTML version deleted]]

___
R-sig-phylo mailing list - R-sig-phylo@r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-phylo
Searchable archive at http://www.mail-archive.com/r-sig-phylo@r-project.org/