Error with diversity beta-group-significance

Technical Support
1

Hi,
I have three samples (BC08, BC09 and BC10). I am trying to run qiime to calculate beta and alpha diversity. However, with the following command it is showing following error.

qiime diversity beta-group-significance
–i-distance-matrix core-metrics-results/unweighted_unifrac_distance_matrix.qza
–m-metadata-file sample-metadata.tsv
–m-metadata-column body-site
–o-visualization core-metrics-results/unweighted-unifrac-body-site-significance.qzv
–p-pairwise

Command…

Plugin error from diversity:

All values in the grouping vector are the same. This method cannot operate on a grouping vector with only a single group of objects (e.g., there are no ‘between’ distances because there is only a single group).

Debug info has been saved to /tmp/qiime2-q2cli-err-lh96uz64.log

Metadata file
BC08 group1
BC09 group2
BC010 group3

2

Good afternoon Manoj,

Here's the main issue:

So it looks like body-site columns of your metadata always has the same values. Can you show me what was inside that column? I don't see it in your example.

Colin

1 Like
3

Hi, I want to see the difference in between the three samples (BC08, BC09, BC10). Here, is my metadata file.

sample-id body-site
#q2:types categorical
BC08 group1
BC09 group2
BC10 group3
4

But if I make any of two samples in groups it produces the output files, however I need to calculate distance in between these three samples separately…

sample-id body-site
#q2:types categorical
BC08 group1
BC09 group1
BC10 group3
5

Ah, OK. Perfect. These distances are already inside of this file:
core-metrics-results/unweighted_unifrac_distance_matrix.qza

You already have the data you need, and don't need to use the
beta-group-significance visualizer at all.

The beta-group-significance visualizer performs a statistical test comparing groups, but you only have one rep in each group, so the statistical test doesn't work. But you can still report the unifrac distances.

Colin

1 Like
7

Ok!
If I need to calculate beta-group-significance visualizer I have replicates samples also. However, the problem is that I can only able to get beta-group-significance bar plot if I put two high read samples in a group either BC08 and BC09 in a group or BC09 and BC10.
----------------------------- metadata file-----------
sample-id body-site
#q2:types categorical
BC08 group1
BC07 group1
BC03 group1
BC01 group1
BC04 group2
BC12 group2
BC11 group2
BC06 group2
BC09 group3
BC02 group3
BC10 group3
BC05 group3

1 Like
8

Thanks for showing me your full metadata. When you run the beta-group-significance visualizer using those 12 samples, does it run successfully?

Are you moving your samples between groups? Have you performed normalization so that the read depth is controlled between samples?

Colin

9

Since I have three samples with high reads, I just want to see the distances in between samples. I have weighted_unifrac_distance_matrix, weighted_unifrac_emperor etc. files.
Can I open in excel or any other software and see the distance between the groups?

10

Yes! There's a guide about exporting .qza and .qzv files over here.

Colin

11

Thank you for your information.