indicspecies package - unable to understand what some specific output means

Hi! I tried searching for this problem, but could not find any relevant information. The R package is indicspecies, I've been using this guide:

In the section "4.2 The function signassoc" of the guide, the R code is:

> prefsign = signassoc(wetland, cluster=groups, alternative = "two.sided", control = how(nperm=199))
> head(prefsign)

which works fine with my data, and gives me this output (when I order by p-value):

> prefsign[order(prefsign$psidak),]
                                     0      1 best     psidak
Acidaminococcaceae              0.9992 0.0009    2 0.00179919
Burkholderiaceae                0.9963 0.0038    2 0.00758556
Desulfovibrionaceae             0.9878 0.0123    2 0.02444871
Coriobacteriaceae               0.9819 0.0182    2 0.03606876
Christensenellaceae             0.0505 0.9496    1 0.09844975
Izimaplasmatales                0.0580 0.9423    1 0.11263600
vadinBE97                       0.9294 0.0708    2 0.13658736
Victivallaceae                  0.9249 0.0758    2 0.14585436
Barnesiellaceae                 0.9191 0.0811    2 0.15562279
Bifidobacteriaceae              0.9190 0.0811    2 0.15562279
Family.XIII                     0.0900 0.9102    1 0.17190000        0.1030 0.8986    1 0.19539100
Peptococcaceae                  0.1101 0.8920    1 0.20807799
uncultured.rumen.bacterium      0.8906 0.1103    2 0.20843391
Atopobiaceae                    0.8891 0.1113    2 0.21021231
Shewanellaceae                  0.9378 0.1142    2 0.21535836

The columns "0" and "1" refer to the metadata I ran the test against, but the numbers in this column don't correspond at all to the abundances or presence-absence occurrences of the taxa, and I can't find any description in the tutorial or on the documentation for "signassoc" (

What do these values mean? They seem to mostly sum to ~1, but sometimes they don't (as in the last taxon in the list I pasted into this post).

Any general comments on the indicspecies package if you have experience with it? Is it robust and reliable?

This will be my first but definitely not my last post on this forum :slight_smile:
Looking forward to learning more, cheers.

Might have found it...


"Returns a matrix of p-values, where species are in rows and groups are in columns. Two additional columns indicate the group with lowest p-value and the p-value for this group after Sidak's correction for multiple testing."

After a night of sleep, this makes more sense, so all the values (except the "best" column) are p-values for associations between the metadata category and the taxon, and the psidak-values are different because they are corrected p-values.

1 Like

It's nice of you to share the solution you found, maybe you can mark your own answer as a solution to your question, that way others with the same problem can more easily find the solution.

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.