# The number of features is normal but the number of observed\_otus is too low?
**URL:** https://forum.qiime2.org/t/the-number-of-features-is-normal-but-the-number-of-observed-otus-is-too-low/11625
**Category:** General Discussion
**Tags:** dada2
**Created:** [September 23, 2019, 1:44am UTC](https://forum.qiime2.org/t/the-number-of-features-is-normal-but-the-number-of-observed-otus-is-too-low/11625 "2019-09-23T01:44:11Z")
**Posts on this page:** 8
**Page:** 2
### Author:  [@ben](https://forum.qiime2.org/u/ben)
#### Post date: [September 25, 2019, 3:54pm UTC](https://forum.qiime2.org/t/the-number-of-features-is-normal-but-the-number-of-observed-otus-is-too-low/11625/21 "2019-09-25T15:54:55Z")
Thanks, I noticed you have lost a lot after merged sequences to non-chimeric.
I think just reviewing this would be interesting (you're losing up to 75% of your reads between merged / non-chimeric.
This may explain why you los resolution. I sometimes lose up to 50% of the reads, but I think that's a lot of reads to lose on the last step.
As you can see, I don't lose a lot of sequences between merged and non-chimeric. See this:
> [@loss of reads after DADA2 as chimeras](https://forum.qiime2.org/t/loss-of-reads-after-dada2-as-chimeras/9503):
>
> Hi, I am loosing around 70% of my reads as chimeras. Is my truncation parameters right ? or is it due to some other issues? [demux.qzv](https://cdck-file-uploads-global.s3.dualstack.us-west-2.amazonaws.com/flex002/uploads/qiime21/original/2X/4/4b826c202032bd1a9b989a8eafd42dde1bf66797.qzv) (291.6 KB) My parameters were qiime dada2 denoise-paired \ --i-demultiplexed-seqs analysis/demux-paired-end.qza --p-trim-left-f 7 --p-trim-left-r 7 --p-trunc-len-f 298 --p-trunc-len-r 256 --o-table table.qza --o-representative-sequences rep-seqs.qza --o-denoising-stats denoising-stats.qza This was the denoising stats [denoising-stats.qzv](https://cdck-file-uploads-global.s3.dualstack.us-west-2.amazonaws.com/flex002/uploads/qiime21/original/2X/2/2d6435fc3fe576c9da7ba67078b5547187c364cb.qzv) (1.2…
Also try here:
> [@Dada2 filtering out \>80% of reads as chimeras!](https://forum.qiime2.org/t/dada2-filtering-out-80-of-reads-as-chimeras/7866/14):
>
> The adapter shouldn't actually be necessary to know as those sequences flank your primers which you do (or at least should) know. So if you trim everything before your forward primer and everything after your RC'ed reverse primer, you will remove all non-biological sequence slight_smile
Ben
---
### Author:  [@ben](https://forum.qiime2.org/u/ben)
#### Post date: [September 25, 2019, 4:20pm UTC](https://forum.qiime2.org/t/the-number-of-features-is-normal-but-the-number-of-observed-otus-is-too-low/11625/22 "2019-09-25T16:20:51Z")
Actually, I just trouble shot this for another lab, here is my code for the V3V4 region:
> qiime dada2 denoise-paired
> --i-demultiplexed-seqs ~/QIIME2\_2\_import/paired-end-demux.qza
> --p-trim-left-f 13
> --p-trim-left-r 13
> --p-trunc-len-f 270
> --p-trunc-len-r 270
> --p-n-threads 0
> --o-table ~/QIIME2\_3\_demux/table\_540.qza
> --o-representative-sequences ~/QIIME2\_3\_demux/rep-seqs\_540.qza
> --o-denoising-stats ~/QIIME2\_3\_demux/denoising-stats\_540.qza
---
### Author:  [@sandro.goforit](https://forum.qiime2.org/u/sandro.goforit)
#### Post date: [September 26, 2019, 9:09am UTC](https://forum.qiime2.org/t/the-number-of-features-is-normal-but-the-number-of-observed-otus-is-too-low/11625/23 "2019-09-26T09:09:36Z")
Hi Ben,
Super thanks for your suggestion:smiley:! Now Im trying to figure out this and I will read everything as you recommended.💪
Best,
Sandro
---
### Author:  [@sandro.goforit](https://forum.qiime2.org/u/sandro.goforit)
#### Post date: [September 26, 2019, 1:40pm UTC](https://forum.qiime2.org/t/the-number-of-features-is-normal-but-the-number-of-observed-otus-is-too-low/11625/24 "2019-09-26T13:40:02Z")
Hi Ben,
Your suggestion helped me sososo much!!! 😊😊 Thank you for finding the problem in my qzv-result. And I finished reading the website you recommended, which helped me to know what I can do for the next step.
Super thanks.💚💚
I will update my result.
Best,
Sandro
---
### Author:  [@ben](https://forum.qiime2.org/u/ben)
#### Post date: [September 26, 2019, 1:44pm UTC](https://forum.qiime2.org/t/the-number-of-features-is-normal-but-the-number-of-observed-otus-is-too-low/11625/25 "2019-09-26T13:44:51Z")
Yes, let me tell you I have run against this same problem, after I fixed it I got better beta-diversity distributions. Could you confirm that your PCOAs look the same or better? Thank you. Ben
---
### Author:  [@Decen](https://forum.qiime2.org/u/Decen)
#### Post date: [September 27, 2019, 1:14am UTC](https://forum.qiime2.org/t/the-number-of-features-is-normal-but-the-number-of-observed-otus-is-too-low/11625/26 "2019-09-27T01:14:27Z")
Hi Yue,
I found someting interesting in your primers for V3-V4 region. Could you please tell me what's the length of your amplicons or the location of the primer pair? what kind of platform did you use for NGS (Miseq or Hiseq or Nova 6000)? Thanks
Best,
Decen
---
### Author:  [@sandro.goforit](https://forum.qiime2.org/u/sandro.goforit)
#### Post date: [September 29, 2019, 2:14am UTC](https://forum.qiime2.org/t/the-number-of-features-is-normal-but-the-number-of-observed-otus-is-too-low/11625/27 "2019-09-29T02:14:47Z")
Hi Decen,
Thank you for your help.😀 Sorry for my late reply.
The length of my amplicons is 460 and the primers are CCTACGGRRBGCASCAGKVRVGAAT(upstream) and GGACTACNVGGGTWTCTAATCC(downstream). The platform is Illumina Miseq. Is there something wrong?🧐
Best,
Yue
---
### Author:  [@Decen](https://forum.qiime2.org/u/Decen)
#### Post date: [September 29, 2019, 5:12am UTC](https://forum.qiime2.org/t/the-number-of-features-is-normal-but-the-number-of-observed-otus-is-too-low/11625/28 "2019-09-29T05:12:56Z")
Hi Yue,
Nothing serious.
It looks they are not "common degenerate primer", when I try to test your primer pair in SILVA "[TestPrime](https://www.arb-silva.de/search/testprime/)", it cannot past. I am using the communal primers, amplicon length ~470bp. If you like, Could you please share how comes for your primer pair? Thanks
Best,
[Previous page](https://forum.qiime2.org/t/the-number-of-features-is-normal-but-the-number-of-observed-otus-is-too-low/11625.md?page=1)