# Too many reads lost in ITSxpress
**URL:** https://forum.qiime2.org/t/too-many-reads-lost-in-itsxpress/25339
**Category:** ITSxpress
**Tags:** q2-itsxpress
**Created:** [January 30, 2023, 8:14am UTC](https://forum.qiime2.org/t/too-many-reads-lost-in-itsxpress/25339 "2023-01-30T08:14:29Z")
**Posts on this page:** 5
**Page:** 1
### Author:  [@sudhir19](https://forum.qiime2.org/u/sudhir19)
#### Post date: [January 30, 2023, 8:14am UTC](https://forum.qiime2.org/t/too-many-reads-lost-in-itsxpress/25339/1 "2023-01-30T08:14:29Z")
Hi everyone,
Sorry for raising the same issue again, but I could not get any responses in my previous post and it was closed in 30 days. Probably because it was posted during new year time.
I am running ITSxpress for trimming ITS1 region for trimming data before classifying with UNITE database.
My workflow looks like this:
1. Quality filtering of data;
2. Removal of primer sequences (cutadapt)
3. Import of paired-end data in QIIME2;
4. Trimming using ITSxpress
In the trimming step I am loosing too much of data (close to 90%), I also ran the data given in tutorial [here](https://forum.qiime2.org/t/q2-itsxpress-a-tutorial-on-a-qiime-2-plugin-to-trim-its-sequences/5780) , where the loss was ~12 %.
Can someone help me in understanding the reason behind this loss. And probably a way to retain more reads for dada2 denoising.
Thanks and regards,
Sudhir
---
### Author:  [@colinbrislawn](https://forum.qiime2.org/u/colinbrislawn)
#### Post date: [January 30, 2023, 6:43pm UTC](https://forum.qiime2.org/t/too-many-reads-lost-in-itsxpress/25339/2 "2023-01-30T18:43:57Z")
Good afternoon,
Thank you for your patience. Many people were off during the new year but we are back now!
> [@sudhir19](#):
>
> In the trimming step I am loosing too much of data (close to 90%)
That is high!
> [@sudhir19](#):
>
> I also ran the data given in the tutorial [here](https://forum.qiime2.org/t/q2-itsxpress-a-tutorial-on-a-qiime-2-plugin-to-trim-its-sequences/5780), where the loss was ~12 %.
That is more expected.
In which of those 4 steps is 90% of the data being lost? Filtering, cutadapt, and ITSxpress could all remove data, so knowing which step is removing too much will help us know what to try next.
---
### Author:  [@sudhir19](https://forum.qiime2.org/u/sudhir19)
#### Post date: [February 14, 2023, 7:41pm UTC](https://forum.qiime2.org/t/too-many-reads-lost-in-itsxpress/25339/3 "2023-02-14T19:41:52Z")
Hi @colinbrislawn,
Thanks for your response. I started with 81,022,719 (paired reads), the initial QC left me with 77,370,389 reads (95.49 % retained), after that, I trimmed the primer sequences and imported data to qiime2 (73,795,023 reads in total), then after trimming in ITSxpress I got 5622889 reads (~7.61 % remaining).
Will it make sense if I don't use ITSxpress and directly try assigning taxonomy on full length UNITE classifier?
---
### Author:  [@colinbrislawn](https://forum.qiime2.org/u/colinbrislawn)
#### Post date: [February 14, 2023, 8:14pm UTC](https://forum.qiime2.org/t/too-many-reads-lost-in-itsxpress/25339/4 "2023-02-14T20:14:54Z")
> [@sudhir19](#):
>
> then after trimming in ITSxpress I got 5622889 reads (~7.61 % remaining).
Well, there's your issue! Something is wrong with ITSexpress 🙃
> [@sudhir19](#):
>
> Will it make sense if I don't use ITSxpress and directly try assigning taxonomy on full length UNITE classifier?
Yeah, try denoising with DADA2 without ITSxpress first.
---
### Author:  [@system](https://forum.qiime2.org/u/system)
#### Post date: [March 18, 2023, 2:15am UTC](https://forum.qiime2.org/t/too-many-reads-lost-in-itsxpress/25339/5 "2023-03-18T02:15:37Z")
This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.