Dada2 Merging Issue

[elmo2420 (Lei)]

My sequences aren't merging and I'm not sure where I am going wrong. For my parameters, I am trying to stay within 140-180 (forward) and 200-215(reverse) positions to be able to achieve an 80% for the input passed filter percentage. I also took into consideration that my quality score shouldn't be below 30 in the 25th percentile. I provided the command lines of my parameters:

qiime dada2 denoise-paired \
  --i-demultiplexed-seqs demultiplexed-sequences.qza \
  --p-trunc-len-f 237 \
  --p-trunc-len-r 175 \
  --p-trim-left-f 5 \
  --p-trim-left-r 5 \
  --p-max-ee-f 2.0 \
  --p-max-ee-r 2.0 \
  --o-representative-sequences rep-seqs-237-175.qza \
  --o-table feature-table-237-175.qza \
  --o-denoising-stats dada2-stats-237-175.qza

Here is my demultiplexed sequences interactive quality plot:

And my dada2 summary results:

One of my questions is if the --p-max-ee parameter could be causing this issue but I'm really sure about it. Any suggestion to fix this issue would be great! Thank you in advance!

[colinbrislawn (Colin J Brislawn)]

Hello Lei,

Welcome to the forums!

I like your approach. Let's get these reads to merge!

@elmo2420

One of my questions is if the --p-max-ee parameter could be causing this issue but I'm really sure about it.

Nope, that setting only changes the filtered reads before merging.

@elmo2420

to be able to achieve an 80% for the input passed filter percentage. I also took into consideration that my quality score shouldn't be below 30 in the 25th percentile.

Sounds good. But because the reads are not merging, get creative.

Because almost no reads merge, this makes me think the reads after trimming are not covering the amplicon you sequenced.

How long is your amplicon? Is it longer than 237+175==412 basepairs long?

[elmo2420 (Lei)]

Hi Colin,

Thank you for your response and the information you provided! The amplicon is 300 bp. I recently tried it again and I changed the --p-trunc-len to:

  --p-trunc-len-f 256 \
  --p-trunc-len-r 202 \

I used the equation --p-trunc-len-f + --p-trunc-len-r - amplicon length and my overlap was 150 bp, which got some of my reads to merge shown in the screenshot

[colinbrislawn (Colin J Brislawn)]

I'm glad to hear that!

Merging a 300 bp amplicon is hard. While keeping ~20% of the reads is unfortunate, looks like most of your samples have >10k reads, which is pretty good.

@elmo2420

I used the equation --p-trunc-len-f + --p-trunc-len-r - amplicon length and my overlap was 150 bp

DADA2 tends to work better with smaller overlaps, so that may be worth trying to get more of your reads to merge.