pcoa-biplot problem.

Hello, I want to draw pcoa biplot follow this tutorial. But I get error at qiime diversity pcoa-biplot command. Error message:

Plugin error from diversity:

  The eigenvectors and the descriptors must describe the same samples.

Debug info has been saved to /tmp/173481.1.all_el7.q/qiime2-q2cli-err-zc8osesj.log

My command:

# get releative frequency data from dada2 output
qiime feature-table relative-frequency \
--i-table table-dada2.qza \
--o-relative-frequency-table frequency-table.qza

# get pcoa result
qiime diversity core-metrics-phylogenetic \
--i-phylogeny rooted-tree.qza \
--i-table table-dada2.qza \
--p-sampling-depth 20000 \
--m-metadata-file sample_metadata.tsv \
--output-dir core-metrics-results

# run pcoa-biplot
qiime diversity pcoa-biplot \
--i-pcoa core-metrics-results/bray_curtis_pcoa_results.qza \
--i-features frequency-table.qza \
--o-biplot biplot/bray_curtis_pcoa_biplot.qza

Where am I wrong? Thanks.

1 Like

Hi @11118,
looks like you are using different feature tables for core-metrics and pcoa-biplot. You need to use the same exact table — currently, you are trying to make a pcoa biplot on the full set of samples, but only providing feature table information for a subset of these samples.

1 Like

Thanks @Nicholas_Bokulich , but I still have troubles here. I wan’t to use the same feature table, but pcoa-biplot command needs RelativeFrequency, core-metrics-phylogenetic can’t use core-metrics-phylogenetic needs Frequency at least. Do you have full biplot demo?

you used command with:

To avoid the error, you need to filter out from your table all samples that have minimum frequency of fetures less than 20000 or you can put a lower p-sampling-depth corresponding to the minimum frequency in your table.

1 Like

A quick note here, you can also just used the rarefied feature table that is output from the core-metrics step. Convert that to relative frequency and you are set.