Taxonomy of Gammaproteobacteria

Preliminaries

This vignette illustrates the most useful functions of yatah.

library(ggtree) # from Bioconductor
library(dplyr)
library(yatah)

Data

For this example, we use data from Zeller et al. (2014). It is the abundances of bacteria present in 199 stool samples.

abundances <- as_tibble(yatah::abundances)
print(abundances, n_extra = 2)
#> # A tibble: 1,585 x 200
#>    lineages `CCIS00146684ST… `CCIS00281083ST… `CCIS02124300ST… `CCIS02379307ST…
#>    <chr>               <dbl>            <dbl>            <dbl>            <dbl>
#>  1 k__Bact…        100.              99.8               96.3           99.9    
#>  2 k__Viru…          0.00697          0.128              3.70           0.00471
#>  3 k__Bact…         66.2             24.6               74.2           45.1    
#>  4 k__Bact…         19.1             74.4               11.9           53.0    
#>  5 k__Bact…         12.1              0.0428             7.22           0.0788 
#>  6 k__Bact…          1.86             0.428              0.765          0.361  
#>  7 k__Bact…          0.758            0.388              2.28           0.985  
#>  8 k__Viru…          0.00697          0.128              3.70           0.00471
#>  9 k__Bact…          0.00155          0.00415            0              0      
#> 10 k__Bact…         62.4             21.7               62.3           44.0    
#> # … with 1,575 more rows, and 195 more variables: `CCIS02856720ST-4-0` <dbl>,
#> #   `CCIS03473770ST-4-0` <dbl>, …

taxonomy <- select(abundances, lineages)
taxonomy
#> # A tibble: 1,585 x 1
#>    lineages                                  
#>    <chr>                                     
#>  1 k__Bacteria                               
#>  2 k__Viruses                                
#>  3 k__Bacteria|p__Firmicutes                 
#>  4 k__Bacteria|p__Bacteroidetes              
#>  5 k__Bacteria|p__Actinobacteria             
#>  6 k__Bacteria|p__Verrucomicrobia            
#>  7 k__Bacteria|p__Proteobacteria             
#>  8 k__Viruses|p__Viruses_noname              
#>  9 k__Bacteria|p__Candidatus_Saccharibacteria
#> 10 k__Bacteria|p__Firmicutes|c__Clostridia   
#> # … with 1,575 more rows

Filtering

Here, we have all the present bacteria at all different ranks. As we are just interested in genera that belong to the Gammaproteobacteria class, we filter() the lineages with is_clade() and is_rank(). The genus name is accessible with last_clade().

gammap_genus <-
  taxonomy %>% 
  filter(is_clade(lineages, "Gammaproteobacteria"),
         is_rank(lineages, "genus")) %>% 
  mutate(genus = last_clade(lineages))
gammap_genus
#> # A tibble: 26 x 2
#>    lineages                                                 genus               
#>    <chr>                                                    <chr>               
#>  1 k__Bacteria|p__Proteobacteria|c__Gammaproteobacteria|o_… Escherichia         
#>  2 k__Bacteria|p__Proteobacteria|c__Gammaproteobacteria|o_… Haemophilus         
#>  3 k__Bacteria|p__Proteobacteria|c__Gammaproteobacteria|o_… Enterobacteriaceae_…
#>  4 k__Bacteria|p__Proteobacteria|c__Gammaproteobacteria|o_… Pseudomonas         
#>  5 k__Bacteria|p__Proteobacteria|c__Gammaproteobacteria|o_… Enterobacter        
#>  6 k__Bacteria|p__Proteobacteria|c__Gammaproteobacteria|o_… Aggregatibacter     
#>  7 k__Bacteria|p__Proteobacteria|c__Gammaproteobacteria|o_… Hafnia              
#>  8 k__Bacteria|p__Proteobacteria|c__Gammaproteobacteria|o_… Actinobacillus      
#>  9 k__Bacteria|p__Proteobacteria|c__Gammaproteobacteria|o_… Sinobacteraceae_unc…
#> 10 k__Bacteria|p__Proteobacteria|c__Gammaproteobacteria|o_… Citrobacter         
#> # … with 16 more rows

Taxonomic table

It is useful to have a taxonomic table. taxtable() do the job.

gammaprot_table <-
  gammap_genus %>% 
  pull(lineages) %>% 
  taxtable()
as_tibble(gammaprot_table)
#> # A tibble: 26 x 6
#>    kingdom  phylum      class         order       family       genus            
#>    <chr>    <chr>       <chr>         <chr>       <chr>        <chr>            
#>  1 Bacteria Proteobact… Gammaproteob… Enterobact… Enterobacte… Escherichia      
#>  2 Bacteria Proteobact… Gammaproteob… Pasteurell… Pasteurella… Haemophilus      
#>  3 Bacteria Proteobact… Gammaproteob… Enterobact… Enterobacte… Enterobacteriace…
#>  4 Bacteria Proteobact… Gammaproteob… Pseudomona… Pseudomonad… Pseudomonas      
#>  5 Bacteria Proteobact… Gammaproteob… Enterobact… Enterobacte… Enterobacter     
#>  6 Bacteria Proteobact… Gammaproteob… Pasteurell… Pasteurella… Aggregatibacter  
#>  7 Bacteria Proteobact… Gammaproteob… Enterobact… Enterobacte… Hafnia           
#>  8 Bacteria Proteobact… Gammaproteob… Pasteurell… Pasteurella… Actinobacillus   
#>  9 Bacteria Proteobact… Gammaproteob… Xanthomona… Sinobactera… Sinobacteraceae_…
#> 10 Bacteria Proteobact… Gammaproteob… Enterobact… Enterobacte… Citrobacter      
#> # … with 16 more rows

Taxonomic tree

To have a tree, use taxtree() with a taxonomic table in input. By default, it collapses ranks with only one subrank.

gammaprot_tree <- taxtree(gammaprot_table)
gammaprot_tree
#> 
#> Phylogenetic tree with 26 tips and 7 internal nodes.
#> 
#> Tip labels:
#>  Escherichia, Enterobacteriaceae_noname, Enterobacter, Hafnia, Citrobacter, Pantoea, ...
#> Node labels:
#>  Gammaproteobacteria, Enterobacteriaceae, Pasteurellaceae, Pseudomonadales, Moraxellaceae, Xanthomonadales, ...
#> 
#> Rooted; includes branch lengths.

Instead of a classical plot, we use ggtree (Yu et al. (2017)) to display the tree.

ggtree(gammaprot_tree) +
  geom_tiplab(hjust = 1, geom = "label") +
  geom_nodelab(hjust = 0, size = 3)