Trim paired sequences with Dada2 drastically reduced the sequence count in Table.qzv? Can anyone help me?

[khaknasheen]

Dear Researchers,

I have got paired-end Illumina (2 x 300) dataset for 16S. I use qiime 2 demux summarize commands. The company has provided me with clean data with no primers or barcodes. Then I use DADA2 denoising by using the following commands
qiime dada2 denoise-paired **

--i-demultiplexed-seqs demux.qza **

--p-trim-left-f 6 **

--p-trim-left-r 6 **

--p-trunc-len-f 300 **

--p-trunc-len-r 300 **

--o-table table.qza **

--o-representative-sequences rep-seqs.qza **

--o-denoising-stats denoising-stats.qza

However, the sequence counts in Table. qzv were very low as compared to demux.qzv. The data demux.qzv and Table.qzv is attached to this message.

Please explain to me why the sequence counts are reducing in Table. qzv ? Am I doing any mistake? I would be thankful if you guide me.

Thanks

[Nicholas_Bokulich]

Hi @khaknasheen,

See the many duplicate posts of this question. Low sequence yields from dada2 are common and usually are due to user error or sequence quality. Most likely your sequences are failing to merge or have very low quality, and you will need to check the dada2 stats output to make sure. See those duplicate forum posts for more details on troubleshooting this problem.

Good luck!

[khaknasheen]

Dear Sir,

Please look at the dada2 stats and give me a good suggestions. The company said no need to trim, as the data is clean, no primers no . Could you suggest me? I need more information, I have read other posts, but my idea is not clear what to do to overcome this issue.

Thanks

[khaknasheen]

Dear Researchers,

The quality of my 2 * 300 bp 16 S Paired-end forward reads was high, so first I used the following commands;

*Qiime-dada2 denoise-paired *

qiime dada2 denoise-paired **

--i-demultiplexed-seqs demux.qza **

--p-trim-left-f 6 **

--p-trim-left-r 6 **

--p-trunc-len-f 300 **

--p-trunc-len-r 300 **

--o-table table300.qza **

--o-representative-sequences rep-seqs300.qza **

--o-denoising-stats denoising-stats300.qza

Using the above commands, the lowest and highest sequence counts in my samples was 784 & 3511 respectively. After getting low sequence count in my samples I trim the forward and reverse reads as follows
*Qiime-dada2 denoise-paired *

qiime dada2 denoise-paired **

--i-demultiplexed-seqs demux.qza **

--p-trim-left-f 10 **

--p-trim-left-r 6 **

--p-trunc-len-f 295 **

--p-trunc-len-r 285 **

--o-table table295285.qza **

--o-representative-sequences rep-seqs295285.qza **

--o-denoising-stats denoising-stats295285.qza

The highest sequence count in my sample 21309 and lowest sequence count were 7215. Now my question is that among the above methods, which is the best according to standard merging and denoising methods. I donot have any idea, Do I need to further trim? I am too much confused. I have attached the demux.qzv file and other two files.

Thanks