I've been using Gneiss to visualize the differential abundances found in my samples. I've seen that balance-taxonomy was discontinued. I tried installing songbird and qurro but for some reason, they require earlier versions of python. I'm using Oracle Virtual Box QIIME2 2021.8. I am not pretty sure how do I get back to early versions but I'm afraid to mess up my program so I did not try. Is there a feasible way for me to determine what taxonomy comprises my balance(s)?
Thank you very much!
EDIT: I decided to install another QIIME2 environment with the last version of QIIME2 with gneiss balance-taxonomy (version 2019.10)
Hi @rmbn - I recommend checking out the conda env create command, which will allow you to install multiple conda environments. That way, you don't need to worry about overwriting your old environments, and you can always toggle back and forth between environments with conda activate
For instance, you can have the following two conda environments
RE Virtual Box, that solution is fine for older Windows versions - but you may want to check out the WSL install instructions if you have a newer computer. From what I have observed, it is much faster than Virtual Box.
Hello @mortonjt ! Awesome! Thanks for the response. I did exactly that and was able to create a separate conda environment for 2019.10 so I can utilize the balance-taxonomy feature while retaining my usual qiime2 environment. I just noticed now that wget also recall a specific version of python (I'm pretty new doing this kind of informatics work, just started around November 2021). I'd definitely get the 2020.8 to put them Songbird and Qurro plugins (I read from your old posts that you recommend those over the one in gneiss)
VB is indeed a pain. I have to run zero-free from time to time to free some storage for my SSD. I have not tried doing WSL but I think it can solve my problem with storage space. Does WSL works exactly like Ubuntu terminal commands?
EDIT: I'd like to ask this somehow unrelated quick clarification in gneiss dendrogram heatmap: does red indicates more abundance and blue indicates less abundance?
Hi @mortonjt thanks! I decided to use Songbird and Qurro instead. It was overwhelming at first because there's a lot to tweak unlike the Gneiss balance-taxonomy where everything is in your face but eventually Songbird and Qurro made more sense now especially if I wanted to focus on interplay between taxa, not to mention better visuals/plots. Although the Gneiss heatmap is a good visual output to see a bird's eye view of the differential log abundance. I don't know if this makes sense but would it be possible to generate a hierarchy artifact from Songbird's differential artifact and make a gneiss heatmap out of it?
Also, I'm using WSL now and it's much better than the Virtual Box so I got rid of it. Pretty much the same except for some few insignificant limitations like the qiime visualization command using terminal doesn't work in WSL.