difference between qiime gneiss dendrogram-heatmap and qiime sample-classifier heatmap


I having some problems understanding the difference between these two. I want to do a heamap with OTUS per subject ( control and patients). Any thoughts?

Thank you!

try qiime feature-table heatmap. the gneiss and sample-classifier heatmap actions are specific to those plugins and are a bit specialized (they require specific types of outputs that are generated by those plugins).

Good luck!


I tried it and i recieved this error:

“Image size of 219610x1415 pixels is too large. It must be less than 2^16 in each direction”

My code was: qiime feature-table heatmap \

--i-table table.qza \                        #this is the able from dada2
   --o-visualization heatmap
--output-dir heatmap

Do you know if this is because i have a lot of features?

Yes too many to display. Filter out low-abundance features first (you probably want to see the most abundant or other select features anyway?)

yes the most abundant. What number is accepted as good? 100 features? Less?

that all depends on your personal definition of abundant, and the overall read count. 100 sounds like a good threshold, but is not very “abundant” if you have billions of total reads! Try filtering at different levels and use qiime feature-table summarize to view how many features remain before generating a heatmap.


Thank you so much for your feedback. I will definitely try that!

1 Like

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.