MiSeq quality of R2 reads

[textux]

Hi!

We are performing a mock community experiment to assess how usefull MiSeq is to analyze our environmental samples. We did 2x300 with v3 chemistry.

So we let a company amlify 16S of different bacterial isolate mock communities and then sequence them.

Now, the R2 reads are much worse than the R1, this is no surprise. But I am not sure if the extent to which they are different is normal. We have 20 sequenced 20 samples and the picture is the same for all the 20 samples (See image: upper row is R1 and lower is R2).

http://picpaste.com/pics/combine_images.1400060891.png

Can you just give me your opinion, is it:
1) normal
2) not very nice, but acceptable
3) not acceptable

Thanks!
Michael

[kmcarr]

Our (limited) experience is that MiSeq v3 PE300 does not work as well for amplicon sequencing as v2 PE250. Your results look similar to the small number of samples we have run in our lab.

[GenoMax]

The answer probably depends on you and may be closer to 2 in the list above. As long as you are able to analyze the data and get the results you want (with the understanding that the raw sequence qualities look the way they do for read 2) you may be able to go forward.

You also may want to consider changing your protocol (http://www.ncbi.nlm.nih.gov/pmc/arti...9-2618-2-6.pdf) to avoid the problem with low nucleotide diversity which ultimately leads to the poor Q-scores you are observing.