phyloseq integration

[nick-youngblut]

Given the challenge of formatting all of the data exactly as required for mbImpute():

• condition
  • vector in the same order as the sample order used for the OTU table
• OTU
  • "wide" matrix format, with column names as taxon IDs and row names as samples
• metadata
  • data.table with rownames that match the sample IDs in the OTU table
• D
  • phylogeny distance matrix in which the row and column orders much match the OTU ID order in the OTU table

...it would be VERY helpful to provide support for phyloseq objects so that the user doesn't have to worry about all of this formatting (and less chance of getting the order wrong)

[fenguoerbian]

Currently in phyloseq, distance is sample-wise, not taxon-wise. But this package requires a taxon-wise distance matrix. I think this is one thing to be worked out.

[ruochenj]

Dear users,

Thank you for your interest in our package, I am working on the integration of the phyloseq. I will let you know when I finish.