Hi！
I am amazed by your professionalism and patience, thank you very much！
I have confirmed that it is V3+V4, so which classifier do I need to train my taxonomic classifiers?
3 Likes
Hi,again.
The 341F and 806R primers I use, the expected amplicon size is (806-341)=465, so can I get the parameters of dada2 from here? For example --p-trunk-len-f 250, --p-trunk-len-r 250, ensure that the overlap area is large enough? Do you have any better suggestions? After I use the following code now, only a small part of the sequence is filtered out when merging, so which is better to set the parameter size or use the following code?
qiime dada2 denoise-paired
--i-demultiplexed-seqs paired-end-demux.qza
--p-trim-left-f 0
--p-trim-left-r 0
--p-trunc-len-f 0
--p-trunc-len-r 0
--p-min-overlap 6
--o-table table.qza
--o-representative-sequences rep-seqs.qza
--o-denoising-stats dada2-stats.qza
--p-n-threads 0
Thanks a lot!
1 Like
Hi again.
Thank you for kind words. The purpose of this forum is to help each other with microbiome analysis.
You can use Silva or Greengenes full-length classifiers, or to train your own classifier specifically for your region (more of work, but should be more robust for your data).
I think you should use a command with disabled trunc parameters, since your quality plot looks good enough to me, so you will keep maximum length of your reads to be safe with overlapping region.
You can proceed as it is (you already run the command above and have satisfying results, right?), or you can decrease --p-min-overlap even more, but it depends on how much of reads are not merging after filtering. No need to do it if you are satisfied with merging stats.
3 Likes
Thank you for your help, you are so kind.
Good luck!
This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.