non-Emperor PCoA in q2

Hi all,

Is there a non-Emperor PCoA option in q2 that is similar to "make_2d_plots.py" in q1? I'm looking for the aesthetic and manipulability (via other programs) that can be performed on the "make_2d_plots.py" output.

Regards,
Tyler

1 Like

Hi @Tyler_Carrier,

emperor does have a 2d plot option (I believe it is in the axis tab, just set the 3rd axis to "none"). Right now there is nothing else in q2 that can give you a q1-like pcoa plot.

Two options, both pretty easy. Export your pcoa coordinates artifact and:

  1. use qiime 1 if you like that aesthetic
  2. plot in R (e.g., with phyloseq)

Hi Nicholas,

What file should be used in Qiime1 to make 2D PCoA?

Regards,
Tyler

The pcoa coordinates artifact (i.e., the same input you use for emperor). But you must export that artifact to use it with q1, R, or any other program.

Hi Nicholas,

I've done the following:

qiime diversity beta-phylogenetic
--i-table Table "post-pair, QC, deblur, rare".qza
--i-phylogeny tree_rooted.qza
--p-metric unweighted_unifrac
--p-variance-adjusted
--o-distance-matrix unwUniFrac_distmatrix.qza

qiime diversity pcoa
--i-distance-matrix unwUniFrac_distmatrix.qza
--p-number-of-dimensions 2
--o-pcoa unweighted_unifrac_PCoA.qza

qiime tools export
--input-path unweighted_unifrac_PCoA.qza
--output-path unweighted_unifrac_PCoA_forQ1

#in qiime1
make_2d_plots.py
-i unweighted_unifrac_PCoA_forQ1/ordination.txt
-o unweighted_unifrac_PCoA_forQ1_2Dplot
-m Lv_DiffFeeding_q2_metadata.txt
-b 'Variable'
--ellipsoid_opacity 0
--scree

This results in "IndexError: index 2 is out of bounds for axis 1 with size 2"

Is there is way around this?/Am I using the correct file for q1?

Regards,
Tyler

Hi @Tyler_Carrier,

Sorry, we cannot provide support for QIIME 1 on this forum.

The format of your coordinates file may be incorrect... here's an idea: instead of exporting as you are, use qiime metadata tabulate --m-metadata-file unweighted_unifrac_PCoA.qza --o-visualization unweighted_unifrac_PCoA.qzv, then view that output and download as a CSV. Try using that as input to the qiime 1 command.

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