yulab-smu / ggtree

:christmas_tree:Visualization and annotation of phylogenetic trees

ggtree: an R package for visualization of phylogenetic trees with their annotation data

‘ggtree’ extends the ‘ggplot2’ plotting system which implemented the grammar of graphics. ‘ggtree’ is designed for visualization and annotation of phylogenetic trees and other tree-like structures with their annotation data.

✍️ Authors

Guangchuang YU

School of Basic Medical Sciences, Southern Medical University

https://yulab-smu.top

If you use ggtree in published research, please cite the most appropriate paper(s) from this list:

1. S Xu, L Li, X Luo, M Chen, W Tang, L Zhan, Z Dai, TT. Lam, Y Guan, G Yu. Ggtree: A serialized data object for visualization of a phylogenetic tree and annotation data. iMeta, 2022, 1(4):e56. doi: 10.1002/imt2.56
2. G Yu. Data Integration, Manipulation and Visualization of Phylogenetic Treess (1st edition). Chapman and Hall/CRC, 2022. doi: 10.1201/9781003279242
3. G Yu. Using ggtree to visualize data on tree-like structures. Current Protocols in Bioinformatics, 2020, 69:e96. doi: 10.1002/cpbi.96
   • Source code and data to reproduce figures in the article
4. G Yu^*, TTY Lam, H Zhu, Y Guan^*. Two methods for mapping and visualizing associated data on phylogeny using ggtree. Molecular Biology and Evolution, 2018, 35(2):3041-3043. doi: 10.1093/molbev/msy194
   • Source code to produce Supplementary Material
5. G Yu, DK Smith, H Zhu, Y Guan, TTY Lam^*. ggtree: an R package for visualization and annotation of phylogenetic trees with their covariates and other associated data. Methods in Ecology and Evolution. 2017, 8(1):28-36. doi: 10.1111/2041-210X.12628

💖 Contributing

We welcome any contributions! By participating in this project you agree to abide by the terms outlined in the Contributor Code of Conduct.

ggtree's Issues

ggtree Error: argument "form" is missing with no default

Hi Guangchuang,

I installed the ggtree package via biocLite("ggtree") on Ubuntu 14.04 with R 3.2.3. When I use your example in the vignette for ggtree():

require(ape)
tr <- rtree(10)
ggtree(tr)

I get an error saying:

"Found more than one class "phylo" in cache; using the first, from namespace 'phyloseq'
Error in ggtree(tr) : argument "form" is missing, with no default"

I have no clue why this occurs, especially since "form" is not an argument in the ggtree() function.

Thanks for your efforts so far, any help would be greatly appreciated.

Cheers,
Chris

ggtree-phenotypicdata

Hi,
I need to draw a tree with phenotypic data, have only quantitative data.
My file is an array.How do I convert my file to this format?
For example

(((((((A: 4, B: 4): 6 C: 5): 8, D: 6): 3, E: 21): 10 ((F 4, G: 12 ): 14

H: 8) 13) 13 ((R 5, J: 2) 30 (K: 11, L: 11): 2): 17): 4, F: 56);

It can make the R using the GGtree?

geom_text() and geom_label()

Hi,

Thanks for the great package! I have two questions:

1. Do you plan to add support for geom_label, which has been added to the dev version of ggplot2? This would be nice for mapping something e.g. an expression value to the background of the tip label (using the fill aesthetic in geom_label()).
2. When I run the iris + ape::bionj() example in the vignette, I get lots of warnings from ggplot2 when I try to add geom_label(aes(label = label)), e.g. Warning message: Removed 148 rows containing missing values (geom_text). However I think all the tip labels are being printed successfully. Do you know what causes this warning?

Thanks again

Tom

> sessionInfo()
R version 3.2.2 (2015-08-14)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Ubuntu 14.04.3 LTS

locale:
 [1] LC_CTYPE=en_AU.UTF-8       LC_NUMERIC=C               LC_TIME=en_AU.UTF-8        LC_COLLATE=en_AU.UTF-8    
 [5] LC_MONETARY=en_AU.UTF-8    LC_MESSAGES=en_AU.UTF-8    LC_PAPER=en_AU.UTF-8       LC_NAME=C                 
 [9] LC_ADDRESS=C               LC_TELEPHONE=C             LC_MEASUREMENT=en_AU.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] ape_3.3            ggplot2_1.0.1.9002 ggtree_1.1.21     

loaded via a namespace (and not attached):
 [1] Rcpp_0.12.0         lattice_0.20-33     IRanges_2.2.7       Biostrings_2.36.4   digest_0.6.8       
 [6] plyr_1.8.3          grid_3.2.2          jsonlite_0.9.17     nlme_3.1-122        gtable_0.1.2       
[11] magrittr_1.5        stats4_3.2.2        scales_0.3.0        stringi_0.5-5       zlibbioc_1.14.0    
[16] reshape2_1.4.1      XVector_0.8.0       rstudioapi_0.3.1    S4Vectors_0.6.5     labeling_0.3       
[21] tools_3.2.2         stringr_1.0.0       munsell_0.4.2       parallel_3.2.2      BiocGenerics_0.14.0
[26] colorspace_1.2-6    gridExtra_2.0.0    

color tree

cls <- list(c1=c("A", "B", "C", "D", "E"),
c2=c("F", "G", "H"),
c3=c("L", "K", "I", "J"),
c4="M")

tree <- groupOTU(tree, cls)
library("colorspace")
ggtree(tree, aes(color=group, linetype=group)) + geom_text(aes(label=label), hjust=-.25) +
scale_color_manual(values=c("black", rainbow_hcl(4))) + theme(legend.position="right")

works well. However, I wanna to design the color for each group, how shall I designate color for each group?
eg: i wanna all the blue color in following group goes to gray.

ggtree::gheatmap() manipulation of scale colors

The arguments low and high are ignored in the gheatmap function when the input is of the type character. This arguments work for numeric values. Is there a chance to set a color scale for characters?

set.seed(1440)
tr <- rtree(20)

df <- data.frame(AA = rnorm(20), BB = rnorm(20))
rownames(df) <- tr$tip.label

p <- ggtree(tr)
gheatmap(p, df, low = "white", high = "black")


df <- data.frame(AA = LETTERS[1:20], BB = letters[1:20])
rownames(df) <- tr$tip.label

## Scale fill is ignored
gheatmap(p, df, low = "white", high = "black")

Thanks for the answer.

R version 3.2.4 (2016-03-10)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows 7 x64 (build 7601) Service Pack 1

locale:
[1] LC_COLLATE=German_Germany.1252  LC_CTYPE=German_Germany.1252    LC_MONETARY=German_Germany.1252 LC_NUMERIC=C                   
[5] LC_TIME=German_Germany.1252    

attached base packages:
[1] stats4    parallel  stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] gridExtra_2.2.1         org.Hs.eg.db_3.2.3      GO.db_3.2.2             clusterProfiler_2.4.3   RSQLite_1.0.0           DBI_0.3.1              
 [7] mygene_1.6.0            GenomicFeatures_1.22.13 AnnotationDbi_1.32.3    Biobase_2.30.0          GenomicRanges_1.22.4    GenomeInfoDb_1.6.3     
[13] IRanges_2.4.8           S4Vectors_0.8.11        BiocGenerics_0.16.1     BiocInstaller_1.20.1    Rcpp_0.12.3             ggtree_1.2.17          
[19] ggplot2_2.1.0           ape_3.4                

loaded via a namespace (and not attached):
 [1] lattice_0.20-33            tidyr_0.4.1                png_0.1-7                  Rsamtools_1.22.0           Biostrings_2.38.4         
 [6] digest_0.6.9               assertthat_0.1             R6_2.1.2                   plyr_1.8.3                 chron_2.3-47              
[11] futile.options_1.0.0       acepack_1.3-3.3            httr_1.1.0                 sqldf_0.4-10               zlibbioc_1.16.0           
[16] lazyeval_0.1.10            curl_0.9.6                 SparseM_1.7                rpart_4.1-10               qvalue_2.2.2              
[21] gsubfn_0.6-6               proto_0.3-10               labeling_0.3               splines_3.2.4              BiocParallel_1.4.3        
[26] stringr_1.0.0              foreign_0.8-66             topGO_2.22.0               igraph_1.0.1               RCurl_1.95-4.8            
[31] biomaRt_2.26.1             munsell_0.4.3              rtracklayer_1.30.3         KEGGREST_1.10.1            nnet_7.3-12               
[36] SummarizedExperiment_1.0.2 Hmisc_3.17-2               XML_3.98-1.4               dplyr_0.4.3                GenomicAlignments_1.6.3   
[41] bitops_1.0-6               grid_3.2.4                 nlme_3.1-125               jsonlite_0.9.19            gtable_0.2.0              
[46] magrittr_1.5               scales_0.4.0               graph_1.48.0               stringi_1.0-1              GOSemSim_1.28.2           
[51] XVector_0.10.0             reshape2_1.4.1             latticeExtra_0.6-28        DO.db_2.9                  futile.logger_1.4.1       
[56] Formula_1.2-1              lambda.r_1.1.7             RColorBrewer_1.1-2         tools_3.2.4                survival_2.38-3           
[61] colorspace_1.2-6           cluster_2.0.3              DOSE_2.8.2                

get_clade_position function

clade is an important concept in phylogenetic tree. We may want to export a selected clade in a very large tree via gzoom function, which works in tree object. If we want to zoom in selected clade of a tree view that was already annotated with ggtree, it's not possible to use with gzoom.

Instead of providing another similar function that works with tree view, I think it would be more flexible to provide the clade position containing xmin, xmax, ymin and ymax. So that user can zoom in the clade by restricting the coordination system to the selected area.

require(ggplot2)
require(ggtree)
p = ggtree(rtree(30)) + geom_tippoint()
cpos = get_clade_position(p, node=45)
p2 = with(cpos, p+xlim(xmin, xmax)+ylim(ymin, ymax))
print(p2+ ggtitle("zoom in selected clade"))

Then we can put them together side by side, just like gzoom.

p1 = p + geom_hilight(node=45)
require(gridExtra)
grid.arrange(p1, p2, ncol=2)

And more flexible to embed tree view using subview function.

with(cpos, subview(p2, p1+theme_transparent(), x=xmin+(xmax-xmin)*.2, y=ymin+(ymax-ymin)*.8))

With the position information, user can annotate clade using geom_segment and geom_text, it will be more flexible than annotation_clade to tune detail.

with(cpos, p1 + geom_segment(x=xmax + .1, xend=xmax+.1, y=ymin, yend=ymax) + 
geom_text(x=xmax+.2, y=(ymin+ymax)/2, label='clade text', hjust=0) + xlim(NA, 7))

collapse and geom_tiplab are incompatible

Here is some example code:

tree <- read.tree(system.file("extdata", "sample.nwk", package="ggtree"))
cp <- ggtree(tree) + geom_tiplab()
cp
cp <- cp %>% collapse(node=21)
cp

It leads to the following errors:

Error in unit(x, default.units) : 'x' and 'units' must have length > 0
In addition: Warning messages:
1: Removed 4 rows containing missing values (geom_segment).
2: Removed 4 rows containing missing values (geom_segment).
3: Removed 14 rows containing missing values (geom_text).

And, no tree is printed. I think that by default the behaviour should be to print out the remaining labels, attached to uncollapsed tips.

I am using ggtree version 1.0.20.

Thank you.

Anders.

ggtree - circular Tree

I am having trouble formatting the names in the circular tree. The names are in the default position.
I need the names remain in the radial position (each name in front of your branch, the radial position).

Script used:
ggtree(tree,layout="circular",branch.length="none",aes(color=group))+ geom_text(aes(label=label, angle=angle), size=2,vjust=-0.3, hjust=1) +scale_color_manual(values=c("slateblue4","firebrick"))

Thanks.

get.offspring.tip() returns null for all nodes

Very useful function in concept, but I couldn't get it to work. Looking at the source code, I have to wonder if it's not because it's basically a wrapper around ape's extract.clade(), which accepts a phylo object, while get.offspring.tip() accepts a ggtree object.

For now I'll probably run get.offspring.df() and filter out the internal nodes.

non overlapping branches for layout="fan" || radial

me again,
if you add a scale_y_continuous after the current line 61 in ggtree, we can have non overlapping branches for those two coord_polar-ized layouts

eg:

p <- p + scale_y_continuous(limits=c(0, sum(p$data$isTip)))

support ancestral sequences inferred by phangorn

## pmlTree is output from 'phangorn::optim.pml'
x=phyPML(pmlTree)
ggtree(x) + geom_text(aes(label=AA_subs), size=2)

Bug in gplot.heatmap

Nice package! I've been looking for something like this for a while. Noticed a bug in gplot.heatmap where the levels of the heatmap aren't reordered properly and the labels don't match up with their respective values in the heatmap.

This line in gplot.heatmap simply reorders the label:

levels(dd$Var1) <- df$label[order(df$y)]

This correctly reorders the levels:

dd$Var1 <- factor(dd$Var1, levels = df$label[order(df$y)])

Also, for very large heatmaps the tree and heatmap don't line up properly. It would be nice to see if this could be fixed because now it requires manually playing with the plot margins until it looks good.

Example:

t = rtree(1000)
data = as.data.frame(replicate(4, rnorm(1000)))
row.names(data) = t$tip.label
gplot(ggtree(t),data)

gheatmap issues

Hi

I am trying to use gheatmap and create a heatmap with just one variable. Its crashing. can you please let me know if it works for you.
Also i am using this code and the error. Can you please tell me what the error means so that i can fix the inputs

genotype1 <- read.table("hh_heatmap.txt", sep="\t", stringsAsFactor=F, header = TRUE)
gheatmap(p, genotype1)
Error in if (any(dd$value == "")) { :
missing value where TRUE/FALSE needed

read nhx tree files

Could you add support for reading NHX formatted files? It is an extended version of the standard newick format where comments with node or branch properties are allowed, eg:
((A[&&NHX:index=1]:1,B[&&NHX:S=2:index=2]:1):1,C[&&NHX:property=35:index=2]:1);
The comment string is included within a [&&NHX: and a ] and has a 'tag=value' format separated by semicolon in the case of multiple tags.

Problem to install

Hi,

any advise why I cannot install ggtree?

devtools::install_github("GuangchuangYu/ggtree")
Downloading github repo GuangchuangYu/ggtree@master
Installing ggtree
Skipping 1 packages not available: Biostrings
'/usr/lib/R/bin/R' --no-site-file --no-environ --no-save --no-restore CMD INSTALL  \
'/tmp/Rtmp7plp5Y/devtools9ab788b5182/GuangchuangYu-ggtree-b82ce5a'  \
--library='/home/paterno/R/x86_64-pc-linux-gnu-library/3.2' --install-tests 

* installing *source* package ‘ggtree’ ...
** R
** inst
** tests
** preparing package for lazy loading
Creating a generic function for ‘nchar’ from package ‘base’ in package ‘S4Vectors’
Error : objeto ‘GeomPoint’ não foi exportado por 'namespace:ggplot2'
ERROR: lazy loading failed for package ‘ggtree’
* removing ‘/home/paterno/R/x86_64-pc-linux-gnu-library/3.2/ggtree’
Erro: Command failed (1)

annotation_clade2 no longer accepts hjust parameter

I upgraded my installation of ggtree today from the bioconductor version to the current github version because I had a special request for the ggtree(ladderize = TRUE) feature, but the upgrade had the unexpected side effect of removing the hjust parameter for annotation_clade2() function.

Eg. I get the error message "Error: Unknown parameters: hjust" under the github version but not the bioconductor version for the following.
ttree <- read.tree(file = "example.tree.txt")
ttree <- midpoint(ttree)
gtree <- ggtree(ttree, ladderize = FALSE)
mytree <- gtree + geom_text(aes(label =label), hjust = -0.03)
annotation_clade2(mytree, tip1 = "C", tip2 = "D", label = "test", angle = 0, hjust = 0)

I still see the error if ladderize = TRUE on the github version.

example.tree.txt

geom_tree2 won't work with aes(color=CategoricalVAR)

library(ggtree)
set.seed(123)
tree=rtree(30)
tree=groupClade(tree, 34)
df=fortify(tree)
ggplot(df, aes(color=x)) + geom_tree2()

Mapping color to continuous variable is OK.

But if I try mapping to categorical variable, it won't work.

ggplot(df, aes(color=group)) + geom_tree2()

The data has no problem.

> head(df)
  node parent branch.length        x y label isTip   branch angle group
1    1     35     0.7989248 2.410756 4   t10  TRUE 2.011294    42     1
2    2     35     0.1218993 1.733731 5   t28  TRUE 1.672781    30     1
3    3     34     0.5609480 1.730580 3   t25  TRUE 1.450106    60     1
4    4     37     0.7533079 2.211171 7   t19  TRUE 1.834517     6     0
5    5     38     0.3744628 2.727372 8   t17  TRUE 2.540140    -6     0
6    6     38     0.6651152 3.018024 9   t29  TRUE 2.685466   -18     0

and it works with original implementation, geom_tree().

ggplot(df, aes(color=group)) + geom_tree()

source code: https://github.com/GuangchuangYu/ggtree/blob/master/R/geom_tree.R

Bioconductor warning with R version 3.1.2

Hi,

Thanks for the nice package. Just successfully installed ggtree via github on R 3.1.2 (devtools::install_github("GuangchuangYu/ggtree")), but was unable to install via Bioconductor:

source("http://bioconductor.org/biocLite.R")
biocLite("ggtree")

Even though the depends on is R >=3.1.0 I get an error message :

Warning message:
package ‘ggtree’ is not available (for R version 3.1.2) 

In case it is helpful I am running Linux Fedora 19 (and yes probably time to update R. :)) and session info when trying to install via Bioconductor is:

R version 3.1.2 (2014-10-31)
Platform: x86_64-redhat-linux-gnu (64-bit)

locale:
 [1] LC_CTYPE=en_CA.UTF-8       LC_NUMERIC=C               LC_TIME=en_CA.UTF-8       
 [4] LC_COLLATE=en_CA.UTF-8     LC_MONETARY=en_CA.UTF-8    LC_MESSAGES=en_CA.UTF-8   
 [7] LC_PAPER=en_CA.UTF-8       LC_NAME=C                  LC_ADDRESS=C              
[10] LC_TELEPHONE=C             LC_MEASUREMENT=en_CA.UTF-8 LC_IDENTIFICATION=C       

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] BiocInstaller_1.16.5

loaded via a namespace (and not attached):
[1] tools_3.1.2

factor handling in geom_tree

hi there,

thanks for your amazing package!
I think that if you turn this line (and possibly for other aes) in geom_tree :
color <- c(color, color) into color <- rep(color, 2) we could directly pass factors to geom_tree (as their column names on the data.frame (or the data.frame resulting from the as.data.frame on the phylo objects). c function converts factors into numeric while rep retains the class.

all the best,
v

Error parsing singleton nodes in rm.singleton.newick

Hi, thanks for the excellent phylogenetic tree annotator.

I am trying to read in BEAST MCC tree files (I have tested several) as newick files, however I first get the error:
Error in read.tree(nwk) :
The tree has apparently singleton node(s): cannot read tree file.
Reading Newick file aborted at tree no. 1

Then applying ggtree:::rm.singleton.newick(nwk,outfile=NULL) results in another error:
Error in ggtree:::rm.singleton.newick(nwk, outfile = NULL) :
can't parse singleton node...

(Note that when I just call rm.singleton.newick, I get the error:
Error: could not find function "rm.singleton.newick")

Any help would be greatly appreciated. Happy to provide further information.

Pinky

Tip did not display properly

Hi Guangchuang,

Thanks for introducing such a nice R package.

I was importing a codeml file. Everything works fine except the tips did not display properly, as the screen shot shows, the tip names were chopped. Am I missing something?

ggtree(mlc, branch.length="none") + geom_text2(aes(subset=!isTip, label=node), hjust=-.3) + geom_tiplab()

Best,
Mao-Lun

FEATURE REQUEST: greater control over parameters for node values

Hello,

I like the flexibility of the package and the ability to work with phylogenetic trees in a graphics oriented way. I think it could be made even better if you could control the display parameters for node labels the way ggtree allows for tip labels. This could include trees passed from other packages. I pruned taxa from a RAxML tree with drop.tip {ape} and labeled clades with ggtree. I would love to exclude values for clades receiving bootstrap values less than 70 percent (or any value...) and move these values to the middle of their branches, leaving the taxa at the tips. I am not expert at R, but I couldn't find an easy way to hack a solution to any of these issues. Still, nice package. I hope it continues to be developed.

Ryan

Add data ring(s) for circular trees

It would be great to have a geom_ring to go with circular trees where at least at first binary data could be plotted (a point if data 1, and blank if 0 or empty). I assume it would use much the similar code used to calculate were to plot tip labels (perhaps something like this: https://medicine.tcd.ie/clinical_microbiology/research/murphy.php).

Ideally, we would be able to add multiple geom_rings to a single ggtree, each offset by a bit.

Or, even more ambitious would be to have a circular ggheatmap function.

Thank you.

Anders.

Could you add some new options?

Hello, thank you for your useful package! It really helps me in visualization of my complex data sets associated with tree!
I often use the visualization tree with associated matrix functionality. But I have some ordinary issues.
I wonder if it is possible to:

1. "Change angle of labels" (column names) in associated matrix. It will be great if you can add this to gheatmap function as a new option.
2. Add option "Show only vertical separating lines" which divide columns in associated matrix. It will be very useful to read complex data (associated matrix) by column manner.
3. Assign a different color scheme for a particular column.

I will be very greatful for this new features!

flip and rotate with unrooted layout

I really like what you have done here!

Not sure if this is a bug or a feature request, but the flip and rotate functions don't seem to work well when layout ='unrooted'. I get overlapping lines. From a quick guess, it looks like the parents are set up differently for these layouts?

Please let me know if you would like any more details, such as an example nwk file and code.

Thanks,
ryan

support ape bootstrap analysis result

> phytree

Phylogenetic tree with 74 tips and 73 internal nodes.

Tip labels:
    Sal_JR3_62_z4_z23_RSK2980, Sal_JR3_62_z36_RKS2983, Sal_JGG_CFSAN001992, Sal_JBX_CFSAN001080, Sal_JM6_CVM19633, Sal_JIX_507440_20, ...

Rooted; includes branch lengths.
## bttree is output from 'ape::boot.phylo'
> bttree
 [1] 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100  80  78
[20]  86 100  73 100 100 100 100 100  79 100  85 100 100 100 100 100 100 100 100
[39] 100 100 100 100 100  99  89 100  94 100 100  73  39 100 100 100 100 100  89
[58] 100 100  73  70  67  97  94 100  74 100  63  46 100 100  89  98
> require(ggtree)
> x <- apeBoot(phytree, bttree)
> x
'apeBoot' S4 object that stored bootstrap value generated by 'ape::boot.phylo' .

...@ tree: 
Phylogenetic tree with 74 tips and 73 internal nodes.

Tip labels:
    Sal_JR3_62_z4_z23_RSK2980, Sal_JR3_62_z36_RKS2983, Sal_JGG_CFSAN001992, Sal_JBX_CFSAN001080, Sal_JM6_CVM19633, Sal_JIX_507440_20, ...

Rooted; includes branch lengths.

> ggtree(x) + geom_label(aes(label=bootstrap))

Aligning tree with other ggplot2

Hi Guangchuang Yu,

First of all thank you for your ggtree package, it is truly versatile and really takes advantage of the good stuff from ggplot2!

I am currently preparing a figure for publication about trait variation and evolution across fiddler crab species. My goal is to have a figure that looks like this: http://ape-package.ird.fr/image/Figure2_TEJEDO.png (Richter-Boix A., Tejedo M. & Rezende E. L. 2011).

Attempt: (attached)
I created a dendogram using ggtree and a ggplot of trait (size) vs species (order). The species name is at the tip of the dendogram. I used ggExtra to arrange them in such a way. While it looks ok, the alignment is not proper.

Request:
Is there a way to align a ggplot like this one with each taxa from ggtree? This will be similar to what has been done with the gheatmap, if we can somehow make the dendogram the x-axis of the ggplot, that would be awesome!

Thank you so much in advance!
Putter

Picture:

Highlight clade function incomplete between tree types

Highlighting multiple clades in a phylogram works fine,

ggtree(tree) %>% hilight(node=134,fill="slategray2",alpha=0.6) %>% hilight(node=82,fill="chocolate2",alpha=0.6) %>% hilight(node=91,fill="darkgoldenrod2",alpha=0.6) ... + geom_text(aes(label=node), size=2, color="purple", hjust=-.2, na.rm = T)

but when I change the layout type to circular, it will not highlight the same clades

ggtree(tree, layout="circular") %>% hilight(node=134,fill="slategray2",alpha=0.6) %>% hilight(node=82,fill="chocolate2",alpha=0.6) %>% hilight(node=91,fill="darkgoldenrod2",alpha=0.6) ... + geom_text(aes(label=node), size=2, color="purple", hjust=-.2, na.rm = T)

Perhaps a problem with my Newick formatting? Strange that it works on phylogram.

ggtree dropping terminal lengths

Hi, thanks so much for this package! It's a million times easier than customizing every tree.

I have a weird issue where ggtree is not plotting a single terminal branch - here's the steps to reproduce:

library(ape)
library(ggtree)
tr<-read.tree(text="(C. sp. 1:0.356,(C. plicata:0.3,((C. guadeloupensis:0.223,((C. portoensis:0.158,C. virilis:0.201):0.005,((C. sp. 8:0.169,(C. angaria:0.031,C. castelli:0.037):0.131):0.073,(C. drosophilae:0.027,C. sp. 2:0.03):0.161):0.047):0.033):0.074,((C. kamaaina:0.14,(C. japonica:0.16,((C. imperialis:0.082,C. afra:0.182):0.009,((C. yunquensis:0.086,C. nouraguensis:0.085):0.038,C. macrosperma:0.159):0.057):0.012):0.059):0.041,(C. elegans:0.166,(((C. brenneri:0.129,C. doughertyi:0.111):0.042,(C. wallacei:0.083,C. tropicalis:0.07):0.033):0.013,(C. remanei:0.129,(C. sp. 5:0.127,(C. briggsae:0.047,C. nigoni:0.032):0.088):0.02):0.017):0.046):0.02):0.195):0.07):0.259);")
plot(tr)
ggtree(tr)+geom_text(aes(label=label),hjust=-0.05,fontface="italic")+geom_text(aes(label=branch.length, x=branch), size = 3, 
              vjust=-0.5, color="#F06C45") +
      scale_x_continuous(breaks=seq(0, 60, 5))

C. imperalis - in the middle is not having its correct length plotted out - the branch length calculator seems to see it fine, but no length is plotted. I've tried deleting and adding branches, and there always seems to be one that is not plotted.

Could not find function "read.raxml"

Hello,

I am trying to read and annotate a raxl tree with bootstrap in ggtree and i get the following error:

raxml_file <- system.file("extdata/RAxML", "/Desktop/standard-RAxML-master/RAxML_bipartitionsBrachLabels_18K_H3.tree", package="ggtree")
raxml <- read.raxml(raxml_file)
Error: could not find function "read.raxml"

Please let me know if there is a solution to this problem.

Thanks

flag to control ladderize

Is it possible to ladderize in the other direction (increasing or decreasing order)?

Thanks!

error reading trees with short branch lengths

Reading trees with read.nhx has some problems with short branch lengths when they are supplied in scientific notation, e.g.:

library(ggtree)
# two identical trees but with the branch length of node A written in scientific notation in the first case
short <- "((A:0.1e-3[&&NHX:Ev=S:S=1:ND=1],B:1[&&NHX:Ev=S:S=1:ND=2]):1[&&NHX:Ev=S:S=1:ND=3],C:1[&&NHX:Ev=S:S=1:ND=4]):1[&&NHX:Ev=S:S=1:ND=5];"
long <- "((A:0.001[&&NHX:Ev=S:S=1:ND=1],B:1[&&NHX:Ev=S:S=1:ND=2]):1[&&NHX:Ev=S:S=1:ND=3],C:1[&&NHX:Ev=S:S=1:ND=4]):1[&&NHX:Ev=S:S=1:ND=5];"
write(short, file = "~/short.tre")
write(long, file = "~/long.tre")
# both are parsed fine by read.tree
plot(read.tree("~/short.tre"))
plot(read.tree("~/long.tre"))
read.nhx("~/long.tre") # works
read.nhx("~/short.tre") # parsing never ends!

image for discussion

hosting images for posting in forum.

Add functionality for ape plot options for "fan" shaped plot - open.angle and rotate.tree

Hi!
It would be great if ggtree would allow you to rotate and open up a fan/circular tree layout like in ape
here is some rough code to plot some examples

library(ape)
library(geiger) # to access whales data
data("whales")
tr = whales$phy # select whales tree
old.par <- par(mfrow=c(1, 3)) # grid plots
plot(tr, show.tip.label = FALSE, type = "f")
title(main = "Orignal")
plot(tr, show.tip.label = FALSE, type = "f", open.angle = 20, rotate.tree = 10)
axisPhylo()
title(main = "Rotate & Open for axis")
plot(tr, show.tip.label = FALSE, type = "f", open.angle = 180)
title(main = "Open to 180 degrees")
par(old.par)

the first plot is the original, the second is where open.angle and rotate.tree are used so axis.phylo() can be used. the third plot is where open.angle is used to plot the tree as a semicircle.

Thank you for your help and for developing/maintaining/expanding this great package!

msaplot with sequence letters?

Not sure if the following can be done currently with ggtree:

I want to annotate a tree with the tip sequences, but I want to see the letters of the sequences rather than just the colors. Right now I do

p <- ggtree(t) + geom_tiplab(size=tip_size) + geom_tippoint(size=3) + geom_treescale()
msaplot(p, "tmp.fasta")

where tmp.fasta is the fasta file. But these lines just show colors and I want the letters. I have short sequences of 30 letters, so visually this would be meaningful.

feature request: phylo4d compatibility

Hi there,

Really liking ggtree - thanks so much for a great tool. One question - have you considered supporting phylo4d objects? Seems like it would be a natural fit for ggtree. As it stands, I seem to have to use data outside of aes() clauses for things like geom_tippoint, as it looks for an object with rows = # of nodes (not just number of tips). Additionally, I have to be wary of the ordering of data matching that of the ordering of the tips of the original tree, etc.

phylo4d, i think, was created to solve these issues. Perhaps there are other ways of going about this that solve these issues without phylo4d... Any thoughts much appreciated. Thanks!

collapse function not working

I keep getting the same error when trying to collapse the tree saying that the node is a tip, even when it isn't. Below is an example using the test file from the vignette.

nwk <- system.file("extdata", "sample.nwk", package="ggtree")
tree <- read.tree(nwk)
ggtree(tree) + geom_text(aes(label=node))
ggtree(tree) + geom_text(aes(label=node)) %>% collapse(node=21)

Error in get.offspring.df(df, node) : input node is a tip...

sessionInfo()
R version 3.2.2 (2015-08-14)
Platform: x86_64-apple-darwin13.4.0 (64-bit)
Running under: OS X 10.11.1 (El Capitan)

locale:
[1] en_US.UTF-8/en_US.UTF-8/en_US.UTF-8/C/en_US.UTF-8/en_US.UTF-8

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] ggtree_1.0.21   ggplot2_1.0.1   knitr_1.11      phytools_0.5-00 maps_3.0.0-2    ape_3.3         readr_0.2.2    

loaded via a namespace (and not attached):
 [1] phangorn_1.99.14        locfit_1.5-9.1          reshape2_1.4.1          splines_3.2.2          
 [5] lattice_0.20-33         colorspace_1.2-6        expm_0.999-0            stats4_3.2.2           
 [9] survival_2.38-3         EBImage_4.10.1          BiocGenerics_0.14.0     jpeg_0.1-8             
[13] plyr_1.8.3              stringr_1.0.0           zlibbioc_1.14.0         Biostrings_2.36.4      
[17] munsell_0.4.2           gtable_0.1.2            mvtnorm_1.0-3           labeling_0.3           
[21] IRanges_2.2.9           parallel_3.2.2          proto_0.3-10            Rcpp_0.12.1            
[25] scales_0.3.0            plotrix_3.6             clusterGeneration_1.3.4 S4Vectors_0.6.6        
[29] scatterplot3d_0.3-36    jsonlite_0.9.17         XVector_0.8.0           abind_1.4-3            
[33] gridExtra_2.0.0         mnormt_1.5-3            png_0.1-7               digest_0.6.8           
[37] stringi_1.0-1           tiff_0.1-5              msm_1.5                 animation_2.4          
[41] numDeriv_2014.2-1       grid_3.2.2              quadprog_1.5-5          tools_3.2.2            
[45] magrittr_1.5            MASS_7.3-44             Matrix_1.2-2            nnls_1.4               
[49] rstudioapi_0.3.1        fftwtools_0.9-7         igraph_1.0.1            nlme_3.1-122           

when I groupOTU my tree some clades are Broken

Hi,yu
my trouble is when I groupout my tree some clades are broken like this,I don't know how to deal with this trouble!as follow

code is
ggtree(tree, aes(color=group),branch.length="none", layout = "circular", size = 0.5)

But if I cant't groupOUT the tree it's normal,view tree as follow

code is
ggtree(tree, branch.length = "none", layout = "circular", size = 0.5)

Do you kown what's wrong with me?

Hide terminal node labels.

Hello,

Thank you for your ggtree package. I have a Bayesian tree with posterior probabilities used for node labels, with the outgroups collapsed to the root.

p <- ggtree(mbtree, layout = "circular", ndigits = 2) %>% collapse(node = 629)  +
    geom_tiplab(aes(label = label, angle=angle), size = 1, color = "steelblue", hjust = -0.1) +
    geom_text(aes(x = branch, label = prob, angle = angle), size = 1, color = "black", vjust = -1.6)

My problem is that this code labels every terminal node with 1. When annotating/labelling, is there a way to differentiate between inner nodes and terminal nodes, or to hide all terminal node annotation?

Attached is an image of what I'm dealing with.

Best,
Nic

Rplot.pdf

using collapse() and geom_cladelabel() together?

Hi! Thanks for the nice package. I was wondering if there is a way to label clades even if some tips have been collapsed? I tried:

require(ape)
set.seed(123)
tr <- rtree(10)
gtr <- ggtree(tr)
anno_gtr <- gtr+ geom_text(aes(label=node)) +geom_tiplab()
collapsed_anno_gtr <- anno_gtr%>% collapse(node=18)
collapsed_anno_gtr+ geom_cladelabel(node=17, "A")

Which gives me the following warnings and does not annotate clade:

Warning messages:
1: Removed 4 rows containing non-finite values (stat_clad_bar). 
2: Removed 4 rows containing non-finite values (stat_clade_text). 
3: Removed 4 rows containing missing values (geom_segment). 
4: Removed 4 rows containing missing values (geom_segment). 
5: Removed 4 rows containing missing values (geom_text). 
6: Removed 1 rows containing missing values (geom_segment). 
7: Removed 1 rows containing missing values (geom_text). 

Thanks!

tip labels being cut off

When viewing trees with ggtree (but not with default tree viewing functionality), my tip labels appear to run off the image. Resizing the image once it is generated doesn't fix this. Is there any way to extend the area on which the tree is plotted such that the labels will fit?

Midpoint rooting tree causes ggtree to fail, reporting multiple roots

After midpoint rooting a tree, ggtree fails, with the getRoot function (called from inside ggtree()) reporting 'multiple roots founded'.

When called on the same tree from outside ggtree, getRoot only returns one root passed the same tree given to ggtree:

require(ape)
require(phangorn)
require(ggtree)

x<-rtree(10) # a random tree; this is handled correctly
getRoot(x) # correctly returns the root
ggtree(x) # succeeds

y<-midpoint(x) # midpoint root x, generating a new tree y using the phangorn midpoint rooting function
getRoot(y) # succeeds, returning one root
ggtree(y) # fails, reporting 'error in getRoot(tr) multiple roots founded (I think this should read 'found')

Is this a bug?

Thank you
David

Error in namespaceExport(ns, exports)

Hi,
Thanks for the nice package.
I think the package installed correctly, but when I require it appears the following message:
Error in namespaceExport(ns, exports) : undefined exports: score, score<-
Could you help me?
Leifor

Feature request: annotate_image() geom

Annotate_image() works well for adding custom icons to tree tips, but I would like to be able to create a ggplot legend for icons that I map to tree tips based on some column in the tree$data data frame. For example, if I create a tree of bacterial isolates, I might map isolate subtype to the tip label color and the source of the isolate to a set of custom icons. I have examined the source code for annotate_image() and geom_tiplab(), but I do not know enough about custom geoms and rasterGrob manipulation to see how to create this hybrid myself.

You might also consider changing annotate_image() so that the offset parameter has an effect when align = FALSE. Setting xmin <- x - width/2 + offset made my trial tree plots look much nicer than having the center of the icon at the end of the branch.

Your work on this package is much appreciated - it's completely changed for the better how I plot trees in R.

Error : object ‘GeomPoint’ is not exported by 'namespace:ggplot2'

Installation via bioconductor is working fine. source('https://bioconductor.org/biocLite.R');biocLite('ggtree') but not in developer's version.

> devtools::install_github('GuangchuangYu/ggtree')
Downloading GitHub repo GuangchuangYu/ggtree@master
Installing ggtree
'/usr/lib64/R/bin/R' --no-site-file --no-environ --no-save --no-restore CMD  \
  INSTALL '/tmp/RtmpQM5w6s/devtools3de51ea1f4a0/GuangchuangYu-ggtree-ea9d661'  \
  --library='/usr/lib64/R/library' --install-tests

* installing *source* package ‘ggtree’ ...
** R
** inst
** tests
** preparing package for lazy loading
Creating a generic function for ‘nchar’ from package ‘base’ in package ‘S4Vectors’
- Error : object ‘GeomPoint’ is not exported by 'namespace:ggplot2'
- ERROR: lazy loading failed for package ‘ggtree’
- * removing ‘/usr/lib64/R/library/ggtree’
- Error: Command failed (1)
- Execution halted

geom_hilight not working with circular layout

tmpTree <- read.tree(text="(t1:0.8578054991,(t5:0.2007951501,(t3:0.2035957272,(t2:0.5625704257,t4:0.1694741752):0.1775545834):0.160806458):0.3601662391);")
ggtree(tmpTree) + geom_hilight(9);
ggtree(tmpTree,layout="circular") + geom_hilight(9);

The geom_hilight works in the rectangular layout, but not the circular layout (no high lighting appears).

Apparently, not every tree have this problem. As in the following example, both layout works

tmpTree <- read.tree(text="(((t2:0.2664971272,t5:0.9326916183):0.5409086382,t1:0.7683936448):0.6352307415,(t3:0.5023543329,t4:0.06688197237):0.6074321754);")
ggtree(tmpTree) + geom_hilight(9)
ggtree(reorder(tmpTree),layout="circular") + geom_hilight(9)

I'm using ggtree_1.2.10

Feature request: zoom function

ggtree is a cool package.

I want to expand the phylogenetic tree.
I'm glad and if there is a function, such as the zoom function of the ape package

Plotting pie charts at nodes base on trait reconstruction

Hi there. Really like this package. I have being trying to figure out if the subview function can be used to annotate nodes with small pie charts based on an ancestral character reconstruction. This type of node label is possible using the ape function node label(pie = results), where results is a matrix with state probabilities at each node.

This would be a great feature to add if it is not already possible with subview!

Tip did not display properly

Hi Guangchuang,

Thanks for providing this awesome R package.

I was trying to display a tree output from PAML, but the tip did not display properly. Below is my code. Am I missing something?

tr2<-ggtree(mlc, branch.length="none") + geom_tiplab()

Thanks,
Mao-Lun

Install in R 3.1.0 fails

The install in R 3.1.0 fails because of a namespace error.

*installing source package ‘ggtree’ ...
*_R
*_inst
*_tests
*_preparing package for lazy loading
Error : object ‘%<>%’ is not exported by 'namespace:magrittr'
ERROR: lazy loading failed for package ‘ggtree’

The installation process fails for both installing directly from github or using a tar-ball. Are there any dependencies that can cause this?