Auspice view: metadata, drop reference strains, combine multiple trees

General
#1

Dear all,

great software. Three minor issues displaying locally generated files

  1. How to add server side csv meta data (date, geographic location, etc.)? The file naming syntax at https://docs.nextstrain.org/projects/auspice/en/latest/introduction/how-to-run.html is not clear to me.

  2. How do I combine several json tree files into one? Could figure out how to display them side by side but not how to combine them. Frequent situation as I have to split large fasta files for memory limits.

  3. How can I suppress the plotting of the reference strains?

Your help is greatly appreciated.

#2

Hi,

How to add server side csv meta data

This isn’t available as it stands. We allow users to drag&drop additional metadata, but this is client-side, not server-side.

More complicated answer: If you control the server, then the data will be handed to the client via /charon/getDataset API requests, and so you could achieve your aim by dynamically injecting metadata into the JSON at this stage. However it seems easier to just add it to the JSON file itself!

How do I combine several json tree files into one?

Are you are asking how to construct independent trees and then combine them such that the resulting phylogeny represents the combined sequence data? I believe this is an unsolved problem.

How can I suppress the plotting of the reference strains?

Do you mean to hide/remove the diversity (entropy) panel? This can be achieved in 3 ways: If you remove “entropy” from the dataset JSON’s metadata.panels array, then it will not be available for display. If you add metadata.displayDefaults.panels = ["tree", "map"] then it will be disabled by default, but can be toggled on by the user; this is the same effect as setting a URL query of ?d=tree,map.

#3

Million thanks, very helpful @james

How to add server side csv meta data

A working JSON example before and after adding date and location metadata at https://jsoneditoronline.org would be very helpful.

How do I combine several json tree files into one?

Big issue here … Due to memory constraints I had to split my fasta input files for nextclade into junks of approx 5000. No issue combining the resulting csv files for downstream analysis but no chance to plot them (even outside auspice). Would need at least the ID of the parent node of a given sequence in the nextclade output to plot it myself…

How can I suppress the plotting of the reference strains?

Solved. Thank you.

#4

Just for posteriority (who is also working with zsh, R and nextclade but not the full nextstrain pipeline): A JSON example is at https://github.com/nextstrain/augur/blob/master/augur/data/schema-export-v2.json which shows the placement of lat, lon, … No solution for combining the JSONs however - so will have to either reconstruct the tree in R using ape::phylo or move everything now to the nextstrain pipeline…