Diversity metrics display wrong number of samples

User Support
1

Hello there,

I did large part of my analysis and everything was great. Then, I was informed that one of the samples was assigned to an incorrect group. I went back to my metadata file and changed "Yes" to "No" in the appropriate categorical column and clicked save. I went back and made a different table.qzv file using the original table.qza (which is generated without association to the metadata file, right?) and the updated metadata file.
Then, I created a new directory for the new diversity metrics (core...corrected) and generated those using my updated metadata table and the original table.qza file which has not changed.
Here is the code:

(qiime2-2018.6) Kims-iMac:qiime2-EHDP-exp2-2018 wetzelmirage$ qiime feature-table summarize \

--i-table table.qza
--o-visualization table.qzv
--m-sample-metadata-file /Users/wetzelmirage/Desktop/Elk\ Hoof\ Metagenomics/EHDP\ 2018\ metadata.tsv
Saved Visualization to: table.qzv
(qiime2-2018.6) Kims-iMac:qiime2-EHDP-exp2-2018 wetzelmirage$ qiime diversity core-metrics-phylogenetic
--i-phylogeny rooted-tree.qza
--i-table table.qza
--p-sampling-depth 65000
--m-metadata-file /Users/wetzelmirage/Desktop/Elk\ Hoof\ Metagenomics/EHDP\ 2018\ metadata.tsv
--output-dir core-metrics-results-corrected
Saved FeatureTable[Frequency] to: core-metrics-results-corrected/rarefied_table.qza
Saved SampleData[AlphaDiversity] % Properties(['phylogenetic']) to: core-metrics-results-corrected/faith_pd_vector.qza
Saved SampleData[AlphaDiversity] to: core-metrics-results-corrected/observed_otus_vector.qza
Saved SampleData[AlphaDiversity] to: core-metrics-results-corrected/shannon_vector.qza
Saved SampleData[AlphaDiversity] to: core-metrics-results-corrected/evenness_vector.qza
Saved DistanceMatrix % Properties(['phylogenetic']) to: core-metrics-results-corrected/unweighted_unifrac_distance_matrix.qza
Saved DistanceMatrix % Properties(['phylogenetic']) to: core-metrics-results-corrected/weighted_unifrac_distance_matrix.qza
Saved DistanceMatrix to: core-metrics-results-corrected/jaccard_distance_matrix.qza
Saved DistanceMatrix to: core-metrics-results-corrected/bray_curtis_distance_matrix.qza
Saved PCoAResults to: core-metrics-results-corrected/unweighted_unifrac_pcoa_results.qza
Saved PCoAResults to: core-metrics-results-corrected/weighted_unifrac_pcoa_results.qza
Saved PCoAResults to: core-metrics-results-corrected/jaccard_pcoa_results.qza
Saved PCoAResults to: core-metrics-results-corrected/bray_curtis_pcoa_results.qza
Saved Visualization to: core-metrics-results-corrected/unweighted_unifrac_emperor.qzv
Saved Visualization to: core-metrics-results-corrected/weighted_unifrac_emperor.qzv
Saved Visualization to: core-metrics-results-corrected/jaccard_emperor.qzv
Saved Visualization to: core-metrics-results-corrected/bray_curtis_emperor.qzv

(qiime2-2018.6) Kims-iMac:qiime2-EHDP-exp2-2018 wetzelmirage$ qiime diversity alpha-group-significance \

--i-alpha-diversity core-metrics-results-corrected/faith_pd_vector.qza
--m-metadata-file /Users/wetzelmirage/Desktop/Elk\ Hoof\ Metagenomics/EHDP\ 2018\ metadata.tsv
--o-visualization core-metrics-results-corrected/faith-pd-group-significance.qzv
Saved Visualization to: core-metrics-results-corrected/faith-pd-group-significance.qzv
(qiime2-2018.6) Kims-iMac:qiime2-EHDP-exp2-2018 wetzelmirage$ qiime diversity alpha-group-significance
--i-alpha-diversity core-metrics-results-corrected/evenness_vector.qza
--m-metadata-file /Users/wetzelmirage/Desktop/Elk\ Hoof\ Metagenomics/EHDP\ 2018\ metadata.tsv
--o-visualization core-metrics-results-corrected/evenness-group-significance.qzv
Saved Visualization to: core-metrics-results-corrected/evenness-group-significance.qzv

(qiime2-2018.6) Kims-iMac:qiime2-EHDP-exp2-2018 wetzelmirage$ qiime diversity beta-group-significance \

--i-distance-matrix core-metrics-results-corrected/unweighted_unifrac_distance_matrix.qza
--m-metadata-file /Users/wetzelmirage/Desktop/Elk\ Hoof\ Metagenomics/EHDP\ 2018\ metadata.tsv
--m-metadata-column ReportedLession
--o-visualization core-metrics-results-corrected/unweighted-unifrac-reported-lession-significance.qzv
--p-pairwise
Saved Visualization to: core-metrics-results-corrected/unweighted-unifrac-reported-lession-significance.qzv

I started looking at Shannon and Unweighted unifrac. I noticed that my sampleID column was ignored and that the number of samples per group (Yes vs No) was not correct. See, I only have 4 individuals in "Yes" group and 6 in the "No" group. Unifrac calculated for 24 and 15, respectively.
image

In Faith and observed OTUs metrics, I do not see the depth plots and only groups (Yes and No) are shown. My original version previous to updating the metadata table showed the depth sampling for individuals too.
I have several questions:
WHY?
Is it because I updated the metadata table (from Yes to No in a single cell?)?
Do I need to start over?
Is it safe to export data? Do my original files remain in place and unchanged after export?

Thank you for your help.

K

3

Correct. Metadata are independent. You can change these any time and they will not impact your feature table! In fact, the metadata are intended to be flexible and mutable.

It is the number of distances, not the number of sample. See this post (and the rest of that thread) for more detail.

It sounds like you just mixed up the action you used here: before you used alpha-rarefaction and now you used alpha_group_significance, but this is not really clear from your question (sharing the old/new files would help clarify, but you can really just check the provenance in the files to make sure you used the same action and version of qiime2 to produce them)

No! Everything I have seen so far is looking fine.

Yes. Exporting creates a copy of the file. Your artifact file remains in place.

Good luck!

1 Like
5

thanks! Looks like the actions were not the same.

1 Like
6

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