Trimming ambiguous ends16S

[mainly.microbe]

Hi all, I am using the Mothur Miseq SOP to run all my 16s sequencing data. We send our sequences to a 3rd party sequencer and the last batch they sent had so many ambiguous sequences that mothur completely cut them out.

After reaching out, they told me to trim the first 4 - 8 bases off the reverse read because it removes the position where the conserved Ns appear at position 3 of the R2. They also told me to shorten R2 overall before merging. Is this possible with trim.seqs or chop.seqs? All the threads ive seen on this forum are for pyrosequencing and when I tried it myself, I do not think I did it correctly. I do not have Oligo files, just fastq.gz R1 and R2 files.

Any help would be amazing

[pschloss]

There’s a chop.seqs function in mothur. BUT… in my experience, trimming sequences before make.contigs results in lower quality data. Can you use an oligos file to remove the conserved Ns in make.contigs?

Pat

[mainly.microbe]

I ended up using trimmomatic in galaxy because I don’t have oligos files for this data.

[pschloss]

They really should be providing you with all of the information - methods, oligos, etc - that are necessary to reproduce the work. This will be expected if you ever publish the data. Besides, it’s making the downstream analysis much harder.

Pat