Prelude

phyloseq is an incredibly useful R package for the organization, analysis, and graphical visualization of sequencing data. Much of it’s ordination-related utility is derived from (or wraps) functions available from the vegan package. Both of these packages have their own strengths and weaknesses. Unfortunately phyloseq does not pull in all of the funcitonality from vegan, including clustering, the whole gambit of standardization and normalization methods, and significance testing (and who can blame them, there are a ton of features!).

At the same time, one should recognize that the plotting features within vegan use the base plot() structure within R. It can be challenging to make a pretty, easily interpretable, plots using vegan, which is especially true as data objects get larger and larger, practically a trueism in today’s age of massive sequencing projects with hundreds of samples and thousands (or tens of thousands) of species/OTUs. phyloseq harnesses ggplot2 to easily make and modify aethetically pleasing plots.

While I do most of my work within phyloseq, I’ve found it very useful to port data objects between the two packages if I want to carry out a specific analysis or take advantage of certain features. Here are some quick and simple functions/approaches that you can use to move data from phyloseq-compatible objects to vegan-compatable objects and back again.

phyloseq to vegan

Analyzing phyloseq objects in vegan requires you to convert them into simpler data structures (dataframes, matricies, etc). Analysis isn’t the only use; you could use vegan to carry out standardization/scaling on metadata (sample_data()) or to carry out some form of tranformation on OTU tables (otu_table()).

1
2
3
4
5
6
7
8
9
10
11
12
13
14

# convert the sample_data() within a phyloseq object to a vegan compatible data object
pssd2veg <- function(physeq) {
  sd <- sample_data(physeq)
  return(as(sd,"data.frame"))
}

# convert the otu_table() within a phyloseq object to a vegan compatible data object
psotu2veg <- function(physeq) {
  OTU <- otu_table(physeq)
  if (taxa_are_rows(OTU)) {
    OTU <- t(OTU)
  }
  return(as(OTU, "matrix"))
}

vegan to phyloseq

How do you get your altered data back into your phyloseq object (or create a new phyloseq object)? Easy!

1
2
3
4

# move an OTU table from vegan to phyloseq  
otu_table(PhyloseqObject) <- otu_table(veganOTUobject, taxa_are_rows=TRUE)  
# move sample data from vegan to phyloseq
sample_data(PhyloseqObject) <- as.data.frame(veganSampleDataObject)