This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Is there a software that can compare phylogenetic trees to each othe (face to face comparison)?

Hi, I'm currently working on several (seven) phylogenetic trees and I want to compare them to each other in the same time. I want to find what part of them is consistent.

Does anyone know a good software?

Thanks for help :)

comparison visualization tree phylogenetics
• 8,490 views • link
updated by Yiyen • Yiyen • written by litinskys • litinskys

2 answers

Here is a solution with ggtree for face to face trees:

library(ggplot2)  
library(ggtree)  
library(cowplot)

set.seed(8455)  
t1 <- rtree(20)  
t2 <- rtree(20)  


T1 <- ggtree(t1 ) +    
  theme_tree2(legend.position='none', plot.margin = unit(c(0,0,0,0),"cm")) +   
  geom_tiplab()   


T2 <- ggtree(t2) +   
  theme_tree2(legend.position='none', plot.margin = unit(c(0,0,0,0),"cm")) +   
  geom_tiplab(hjust =1) +   
  scale_x_reverse()     


d1 = T1$data[T1$data$isTip,]  
d1$x[] = 1  
d2 = T2$data[T2$data$isTip,]  
d2$x[] = 2  

TTcon <- rbind(d1, d2)  

T1 = ggtree(t1 ) +    
  theme_tree2(legend.position='none', plot.margin = unit(c(0,0,0,0),"cm")) +   
  geom_tiplab(align = TRUE) # if required +   xlim(0,5.2)  
T2 = ggtree(t2) +   
  theme_tree2(legend.position='none', plot.margin = unit(c(0,0,0,0),"cm")) +   
  geom_tiplab(hjust =1, align = TRUE) +   
  scale_x_reverse() # if required add within scale_x_reverse() -> limits = c(5.2, 0))  

L1 = ggplot(TTcon, aes(x = x, y = y, colour = label, group = label)) + geom_line() +   
  theme_void() + theme(legend.position="none", plot.margin = unit(c(1,0,1,0),"cm"))  

cowplot::plot_grid(T1, L1 ,T2, nrow = 1, align = "hv")
• 0 views • link
updated by GenoMax GenoMax • written by ron • ron

Thanks for the code! However, I find that my lines are not fully aligned. I figure this is because one tree is missing branches in the other, also shown in this example:

t1 <- rtree(50)  
t2 <- rtree(20)

T1 <- ggtree(t1 ) +    
  theme_tree2(legend.position='none', plot.margin = unit(c(0,0,0,0),"cm")) +   
  geom_tiplab()   


T2 <- ggtree(t2) +   
  theme_tree2(legend.position='none', plot.margin = unit(c(0,0,0,0),"cm")) +   
  geom_tiplab(hjust =1) +   
  scale_x_reverse()     


d1 = T1$data[T1$data$isTip,]  
d1$x[] = 1  
d2 = T2$data[T2$data$isTip,]  
d2$x[] = 2  

TTcon <- rbind(d1, d2)  

T1 = ggtree(t1 ) +    
  theme_tree2(legend.position='none', plot.margin = unit(c(0,0,0,0),"cm")) +   
  geom_tiplab(align = TRUE) # if required +   xlim(0,5.2)  
T2 = ggtree(t2) +   
  theme_tree2(legend.position='none', plot.margin = unit(c(0,0,0,0),"cm")) +   
  geom_tiplab(hjust =1, align = TRUE) +   
  scale_x_reverse() # if required add within scale_x_reverse() -> limits = c(5.2, 0))  

L1 = ggplot(TTcon, aes(x = x, y = y, colour = label, group = label)) + geom_line() +   
  theme_void() + theme(legend.position="none", plot.margin = unit(c(1,0,1,0),"cm"))  

cowplot::plot_grid(T1, L1 ,T2, nrow = 1, align = "hv")

enter image description here

Is there a way to account for that and only have lines drawn for the matching tip labels?

• 0 views • link
written by Morgan S. • Morgan S.

hello, I have same problem, but I found this is because the padding beteween figure and x axis in line chart. you can chage this by follow this answer on stackoverflow: https://stackoverflow.com/a/52318834.

but I still can't let each line align to each tip perfectly:)

• 0 views • link
written by Yiyen • Yiyen

You can do tree comparison with ete2, although I am not sure it is in the way you meant: https://pythonhosted.org/ete2/tutorial/tutorial_trees.html#comparing-trees

• 0 views • link
written by Giovanni M Dall'Olio Giovanni M Dall'Olio

Thank you for your response.

The link you sent refers to compering between only 2 trees. I am looking for a software which shall allow me to compare all 7 trees together, in the same time.

Is there anything you may recommend?

• 0 views • link
updated by Ram Ram • written by litinskys • litinskys

I don't know of anything that would do multiple trees. You may have to do multiple pairwise comparisons or compare each tree to some reference tree structure.

• 0 views • link
updated by Ram Ram • written by Jautis Jautis

Log in to answer this question.

More posts like this