Your thoughts regarding fungal OTUs

Hi Everyone i have run through QIIME2 with some fungi data and after using DADA2 and not trimming because of the variability of the ITS region I filtered by quality. I hardly lose any read (around 20-30%). I end up with around 1700 OTUs after filtering out the unclassified fungi etc. But the observed diversity between treatments is fairly low , just over 100 species? The ecosystem i am studying is peatlands.

Is it possible that due to the anaerobic nature of peatlands there just isn't many species there?

Best wishes

Hello Martyn,

There is a bunch neat questions here to unpack, and they are all related in different ways!

It's fully possible for a specific ecosystem to be composed of a few primary microbes, and have a low overall richness. Low richness values can also be caused by low sequencing depth, which is a technical issue. Did your alpha rarefaction curves level off, implying sufficient read depth?

I presume there's at least some oxygen near the surface of the peat, and then it reduces the farther down you go resulting in a oxygen gradient. At what depth(s) did you take samples? Is there an O2 gradient or is peat basicly anaerobic at all depths?

We can talk more about DADA2 settings for the ITS region, but I wanted to start with the biological question of species richness.

Let me know what you think!
Colin

1 Like

Hi Colin, thanks again for getting back to me. I went as far down as 60cm. I wonder if the OTUs are collapsing into species and that's why my OTU table is fairly large but the richness is low. For DADA2 i did not trim and filtered only by quality using trun-q because the ITS region is variable although as you know i am very inexperienced.

best wishes.

For ITS I would trim and truncate just like 16S, but try to give extra room to allow the longer regions to overlap.

Also :point_down:

Hi Colin , the rarefaction curves actually plateau very early on so i'm not sure what the problem is.

I am also unsure what parameters to use for the extract classifier for ITS. My primers are ITS1 and ITS2.

qiime feature-classifier extract-reads
--i-sequences unite-ver8-seqs_99_10.05.2021.qza
--p-f-primer CTTGGTCATTTAGAGGAAGTAA
--p-r-primer GCTGCGTTCTTCATCGATGC
--p-min-length ??
--p-max-length ??
--o-reads UNITE.ref-seqs.2.qza

Best wishes

That's good! It looks like your communities might not be very rich after all...

You can read more about these settings in the feature-classifier extract-reads docs.

What is the shortest and longest you expect your amplicon to be? (This is a biological question, not a technical question.)

1 Like

Hi Colin, that's a relief. Thanks again for getting back to me on all my questions.

According the the Earthmicrobiome page the expected amplicon size can be ~250–600 bp which seems a big variation.

Best wishes