Tanglegram and color_lines

[TreeT2]

Hello Tal and All,

I am looking to produce a tanglegram of two phylogenies of the same taxa with different datasets. I have already converted the phylogenies to dendrograms using DECIPHER.

I create an ordered character vector of colours based on tip label order on the left side phylogeny (I also reverse this as color_lines works from bottom to top). My problem is that the colours are not mapping accurately. I do think this is either due to the topology of one phylogeny/dendrogram or using the original phylogeny to generate the ordered vector (although I suspect not).

(note I have also tried ape cophyloplot and experience the same issue)

command : tanglegram(dendA, dendC, color_lines = x1)

Alternatively is there a method to colour associations lines similar to common_subtrees_color_lines, but involves specifying the subtrees/ clade in the left phylogeny (via the scale bar)??

Hope this makes sense.

[talgalili]

Hi there, For future people - how did you solve this?

[TreeT2]

Hi Tal,

I originally posted on stack exchange

The issue was in taxa names, some were labeled as name_101, the underscore did translate in the vector so I was left with name 101. So when I went forth to create a colour vector using extracted taxa names numerous rows were missing. I didn't get an error message so it took me a while to work it back.

[talgalili]

Thanks. What type of test in the code would have been able to pick it up? (If you have an idea for one then we could add it)

[TreeT2]

Im just referring to info in the help page:

a vector of colors for the lines connected the labels. If the colors are shorter than the number of labels, they are recycled (and a warning is issued). The colors in the vector are applied on the lines from the bottom up.

Im afraid Im only a novice in R.

[talgalili]

So are you saying that you do not see a warning message? Which function is this the help file of?

[TreeT2]

No warning message. The quote above is from the tanglegram function and color_line argument.

[virologist]

Hi, Tal I have the same issue. Here is the code which could reproduce the color issue.

rm(list=ls())
library(dendextend)
library(dplyr)
library(DECIPHER)
library(ggplot2)

#load source data
dend1 <- ReadDendrogram('https://raw.githubusercontent.com/virologist/Flu/master/tree1.newick', internalLabels = FALSE)
dend2 <- ReadDendrogram('https://raw.githubusercontent.com/virologist/Flu/master/tree2.newick', internalLabels = FALSE)

#read the clade/color data
tip_metadata <- read.csv('https://raw.githubusercontent.com/virologist/Flu/master/Cluster.csv', sep=",", header=TRUE,check.names=FALSE, stringsAsFactor=F)

#set label colors trees
virus_lab <- labels(dend1)
virus_lab <- tip_metadata$Clade
virus_lab <- as.factor(virus_lab)
levels(virus_lab) <- 1:length(levels(virus_lab))
labels_colors(dend1) <- virus_lab

# Stepwise untangle one tree compared to another
dend2_corrected <- untangle_step_rotate_1side(dend1, dend2)[[1]]

#Create a dendrogram with color branches) 
dl <- dendlist(color_branches(dend1, clusters = as.numeric(virus_lab),col=levels(virus_lab)), 
               color_branches(dend2_corrected, clusters = as.numeric(virus_lab), col=levels(virus_lab)))


#Print entanglement
dl %>% entanglement # lower is bette

dend <- tanglegram(dl, 
           main_left = "HA", 
           main_right = "NA",
           color_lines=as.numeric(virus_lab),
           margin_inner = 0.5,  ## remove tip labels from tanglegram
           lwd = .5, 
           axes = TRUE,
           type = "r",
           remove_nodePar = TRUE, 
           highlight_distinct_edges = FALSE, 
           highlight_branches_lwd=FALSE, 
           common_subtrees_color_lines = FALSE, 
           edge.lwd = .8,
           lab.cex = .1, 
           faster = FALSE)

Best, Yan

[talgalili]

Any chance for a smaller example that is self contained and that can reproduce the issue? Maybe use dput to extract the objects.

Thanks!

On Sun, Jun 14, 2020, 07:27 Biopig [email protected] wrote:

Hi, Tal I think I meet the similar problem. I demo code is fine. The branch and tanglegram link lines have the as same color as I want. [image: image] https://user-images.githubusercontent.com/54756814/84584642-63dcb900-ae39-11ea-83bc-3e6d1a2ecbbf.png Here is the link of all code and files https://github.com/virologist/Flu Here is the code `#if (!require(dendextend)) install.packages(dendextend); rm(list=ls()) setwd("C:/Users/Administrator/Desktop/dendextend_demo")

library(dendextend) library(dplyr) library(DECIPHER) library(ggplot2)

#load source data dend1 <- ReadDendrogram("dend1.newick", internalLabels = FALSE) dend2 <- ReadDendrogram("dend2.newick", internalLabels = FALSE)

#read the clade/color data tip_metadata <- read.csv("taxon_color.csv", sep="\t", header=TRUE,check.names=FALSE, stringsAsFactor=F)

#set label colors trees virus_lab <- labels(dend1) virus_lab <- tip_metadata$clade virus_lab <- as.factor(virus_lab) levels(virus_lab) <- 1:length(levels(virus_lab)) labels_colors(dend1) <- virus_lab Stepwise untangle one tree compared to another

dend2_corrected <- untangle_step_rotate_1side(dend1, dend2)[[1]]

#Create a dendrogram with color branches) dl <- dendlist(color_branches(dend1, clusters = as.numeric(virus_lab),col=levels(virus_lab)), color_branches(dend2_corrected, clusters = as.numeric(virus_lab), col=levels(virus_lab)))

#Print entanglement dl %>% entanglement # lower is bette

tanglegram(dl, main_left = "Tree_1", main_right = "Tree_2", color_lines = as.numeric(virus_lab), lwd = .5, axes = TRUE, type = "r", remove_nodePar = TRUE, highlight_distinct_edges = FALSE, highlight_branches_lwd=FALSE, common_subtrees_color_lines = FALSE, edge.lwd = .8, lab.cex = .1, faster = FALSE) `

BUT, not with my own data.

1. I have 11 clades. But, it looks like the dendextend only can offer 9 colors. The colors of Green and Black have been recycled without any warming. [image: image] https://user-images.githubusercontent.com/54756814/84584517-0136ed80-ae38-11ea-8192-cedd75cdfca0.png
2. The branch color and color_line are not consistent even I designated them coloring based on same taxon.

Could you please give some help? Best Regards Yan

— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub https://github.com/talgalili/dendextend/issues/47#issuecomment-643716880, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAHOJBWQTWCTRFSRPHUT5D3RWRGUTANCNFSM4DQ76FJA .

[virologist]

Yes, It's reproducible.

[talgalili]

In order to help me. Could you please replace: "Here is the link of all code and files https://github.com/virologist/Flu Here is the code `#if (!require(dendextend)) install.packages(dendextend); rm(list=ls()) setwd("C:/Users/Administrator/Desktop/dendextend_demo")"

With something I can just copy paste to my console, run, and see the same results? (i.e.: without sending me to download files to somewhere, change the path, etc.?)

On Sun, Jun 14, 2020 at 8:22 AM Biopig [email protected] wrote:

Yes, It's reproducible.

— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub https://github.com/talgalili/dendextend/issues/47#issuecomment-643720414, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAHOJBRBXS53GZ3KGS6QMALRWRNDFANCNFSM4DQ76FJA .

[virologist]

In order to help me. Could you please replace: "Here is the link of all code and files https://github.com/virologist/Flu Here is the code `#if (!require(dendextend)) install.packages(dendextend); rm(list=ls()) setwd("C:/Users/Administrator/Desktop/dendextend_demo")" With something I can just copy paste to my console, run, and see the same results? (i.e.: without sending me to download files to somewhere, change the path, etc.?) … On Sun, Jun 14, 2020 at 8:22 AM Biopig @.***> wrote: Yes, It's reproducible. — You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub <#47 (comment)>, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAHOJBRBXS53GZ3KGS6QMALRWRNDFANCNFSM4DQ76FJA .

Yes, for sure. I updated the code so that you can copy, paste and run. Yan

[ghoresh11]

Old thread, but was struggling with this today and figured it out.

The colour order shouldn't be the reverse of tree$tip.label, but the reverse of tree$tip.label[order.dendrogram(x = as.dendrogram(tree))]

[talgalili]

Cool, thanks for letting us know! Is there a code change you think we should make in the package?

On Thu, Jun 3, 2021 at 1:44 PM Gal Horesh @.***> wrote:

Old thread, but was struggling with this today and figured it out.

The colour order shouldn't be the reverse of tree$tip.label, but the reverse of tree$tip.label[order.dendrogram(x = as.dendrogram(tree))]

— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub https://github.com/talgalili/dendextend/issues/47#issuecomment-853774529, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAHOJBXH553VAW3WMAILWO3TQ5MI3ANCNFSM4DQ76FJA .

[ghoresh11]

The details of the code are not so clear to me on a quick look, but I realised that the order of the lines came down to this line.

  ord_arrow <- cbind((1:l)[order(order.dendrogram(dend1))], (1:l)[order(order.dendrogram(dend2))])

Perhaps if this line is changed to the order is based on a match between dend1$tip.label and dend2$tip.label it will solve the problem.

[talgalili]

Any chance you could make a simple small reproducible example of the issues so I can debug it?

On Thu, 3 Jun 2021, 13:57 Gal Horesh, @.***> wrote:

The details of the code are not so clear to me on a quick look, but I realised that the order of the lines came down to this line.

ord_arrow <- cbind((1:l)[order(order.dendrogram(dend1))], (1:l)[order(order.dendrogram(dend2))])

Perhaps if this line is changed to the order is based on a match between dend1$tip.label and dend2$tip.label it will solve the problem.

— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub https://github.com/talgalili/dendextend/issues/47#issuecomment-853781359, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAHOJBTXXAOB5WGLE2UPD2TTQ5NZNANCNFSM4DQ76FJA .

[ghoresh11]

This should help:

library(ape)
library(dendextend)
library(phytools)

tree1 = rtree(n = 20)
tree2 = rtree(n = 20)

## the problem arises when you reroot the tree and the order changes
tree1 = chronos(root(tree1, "t17"))
tree2 = chronos(root(tree2, "t17"))

## order based on tips
colours = data.frame(tip = tree1$tip.label, colours = rep("", length(tree1$tip.label)))
colours$colours[colours$tip %in% c("t1","t2","t3","t4","t5")] = "red"
colours$colours[colours$tip %in% c("t6","t7","t8","t9","t10")] = "blue"
colours$colours[colours$tip %in% c("t11","t12","t13","t14","t15")] = "purple"
colours$colours[colours$tip %in% c("t16","t17","t18","t19","t20")] = "orange"



## this is wrong, you expect 1-5 to have the same colout, 6-10, 11-15 and 16-20, but this isn't the case
tanglegram(tree1, tree2, intersecting  = F,
           axes = F,highlight_branches_lwd = F, color_lines = colours$colours,
           dLeaf = 0.01)

correct_order = order.dendrogram(x = as.dendrogram(tree1))
colours = colours[correct_order,]


tanglegram(tree1, tree2, intersecting  = F,
           axes = F,highlight_branches_lwd = F, color_lines = colours$colours,
           dLeaf = 0.01)