Observed Species calculation

User Support
1

Thanks! This is a super useful resource.

Additional question:
Does anyone know which I should use to calculate alpha diversity the number of observed species (this was an option in QIIME but I don’t know which it is in QIIME2)? Or how I would go about calculating the number of observed_species in QIIME2?

Thanks!

2

Hi @LVC,

This has had a lot of names and meaning over hte years! The observed_otus metric should maybe be called observed features instead! For observed species, you need to collpase your table to species level using qiime feature-table group with your taxonomy as metadata, and then use the observed_otus for your alpha diversity metric.

(However, because I like to be that person, let me ask: do you trust your species designations?)

Best,
Justine

1 Like
3

Thanks Justine,

observed_species isn't one of my options when I type:
qiime diversity alpha --help
which is what is confusing me...or should I be using a different command to qiime diversity alpha for this?

I have osd and observed_otus as options though. I am looking for a richness indicator that I can convert into the diversity hill number system (observed_species preferably for a q = 0 (hill number calculation) so I can plot some graphs to look at diversity. q = 0 would be observed species, shannon (q = 1), simpson (q = 2) .... berger parker (q = infinity).

It's okay, I've already thought about the species designations aspect

4

Hi @LVC,

No, observed species and observed otus are the same metric, whether you choose to call them species or ASVs, the difference is how you choose to represent your data. So, if you want to get species level, you collapse to species level and then you use --metric observed_otus because that will count the number of unique features (in htis case, species because you’ve collapsed your table) in each sample.

Best,
Justine

1 Like
5

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