ITSexpress and ITS data upstream analysis

[shaista_karim]

Hello,

I am looking at some ITS data. I am hitting some roadblocks, any recommendation will be highly helpful.

I successfully looked into 16S data following the available tutorials and some help from the QIIME2 forum to fix some bugs and now I switched to ITS data. I have demultiplexed paired data.

I understand the fact that ITS and 16S data analysis will be similar but upstream analysis is different.
I am aware of previous tutorials "Q2-ITSxpress: A tutorial on a QIIME 2 plugin to trim ITS sequences" I tried to trim my data using itsecpress but after trimming I am left with almost no features.
I used code "qiime itsxpress trim-pair --i-per-sample-sequences demux-paired-end.qza --p-region ITS2 --p-taxa F --p-threads 12 --o-trimmed itsexpress-trimmed.qza"

demux-paired-end.qzv (338.4 KB)

itsexpress-trimmed.qzv (313.5 KB)

I tried using DATA2 to trim and it did work fine.
dada2-stats-summ.qzv (1.2 MB)

I then tried the previously available classifier and got the following output

I am aware of available resources to download the latest version of classifier "UNITE - Resources"

My questions are:

1. If I am able to trim my data using DATA2 do I still need itsExpress to trim my files? I read about on the forum but still not sure I fully understand.
2. Is there any updated tutorial available for ITS analysis in QIIME 2 docs that are compatible with QIIME2 2021 release?
3. Unfortunately, I am unable to download the UNITE files from this site- any idea how possibly can I get access to the files "UNITE - Resources".

[colinbrislawn (Colin J Brislawn)]

It's good to hear from you again. Sounds like your Dockerized Galaxy deployment of Qiime2 is working well.

I'm not an expert on ITSxpress, but it looks like something went very wrong. Not only are all by 17 of your reads missing, the whole reverse read is gone!

It wonder if passing a different region other than --p-region ITS2 would help. Are you sure these reads came from ITS2?

---

Your dada2 stats look pretty good!

@shaista_karim

• If I am able to trim my data using DATA2 do I still need itsExpress to trim my files? I read about on the forum but still not sure I fully understand.

@benjjneb

In short, DADA2 does not recommend processing pre-merged reads, because the different regions of pre-merged reads (Forward-only, overlapping, reverse-only) have different relationships between the assigned quality scores and the error rates, which can lead to false-positive ASV inference. You can see more discussion of this here: https://github.com/benjjneb/dada2/issues/327#issuecomment-400022629

That's tutorial is the most up to date that I can find too. There is also documentation on the plugin page and on Github.

Here are direct links to the Unite database files for version 8.3:

# 8.3	2021-05-10	Fungi	14 097	44 343	Current	https://doi.org/10.15156/BIO/1264708
wget -qO- https://files.plutof.ut.ee/public/orig/C5/54/C5547B97AAA979E45F79DC4C8C4B12113389343D7588716B5AD330F8BDB300C9.tgz | tar xz -C downloads --strip-components 2 # sh_qiime_release_10.05.2021       # normal

# 8.3	2021-05-10	Fungi	14 097	83 993	Current	https://doi.org/10.15156/BIO/1264763
wget -qO- https://files.plutof.ut.ee/public/orig/B3/05/B3054DF783AC61A0C3BD0FDEB0516EC394934809AAE43CA0F3081C0A184FAA39.tgz | tar xz -C downloads --strip-components 2 # sh_qiime_release_s_10.05.2021     # add s for 97% singletons

# 8.3	2021-05-10	All eukaryotes	14 237	96 423	Current	https://doi.org/10.15156/BIO/1264819
wget -qO- https://files.plutof.ut.ee/public/orig/48/29/4829D91F763E20F0F4376A60AA53FC9FBE6029A7D1BDC1B45347DD64EDE5D560.tgz | tar xz -C downloads --strip-components 2 # sh_qiime_release_all_10.05.2021   # add all for Euks

# 8.3	2021-05-10	All eukaryotes	14 237	190 888	Current	https://doi.org/10.15156/BIO/1264861
wget -qO- https://files.plutof.ut.ee/public/orig/1D/31/1D31FA3A308BDC2FB2750D62C0AA40C5058C15405A3CC5C626CC3A3F5E3903ED.tgz | tar xz -C downloads --strip-components 2 # sh_qiime_release_s_all_10.05.2021 # and s and all for 97% Euks singletons

I have pre-built versions of the Unite classifiers available over here, if you would like to give them a try.

[shaista_karim]

Thank you so much @colinbrislawn - I tried both Unite version 8.3 and the pre-built version of the Unite classifier you shared worked-- Yayy