I have two .fastq files for all of my samples. I have one mapping file for each of those (2 mapping files). Two samples (one from each fastq file/mapping file) share a barcode sequence.
I ran through the moving pictures tutorial to the dada2 step for both of the files.
Would I be able to concatenate the resultant files (rep_seqs & table) for the two and go through the taxonomic assignment using the combined file?
Thank you,
Megan
Hello Megan,
I think you can combine these data sets, but I would merge before you have the rep_seqs table. The main strategy here is to qiime tools import these two runs separately, and make sure all samples have unique names. After that, you can denoise then merge (or merge then denoise) your data set, because now all the sample names are unique.
Combining tables later on is risky because matching OTUs or samples names could be joined incorrectly, leading to wonky results. When it comes to merging, sooner is safer.
I hope that helps,
Colin
P.S. This tutorial includes a section on the denoise and merge strategy.
Hi @megladds and @colinbrislawn,
For denoising with DADA2, the DADA2 developers recommend that denoising be done on a per-run basis, so we also recommend that you merge after denoising (not before denoising). As @colinbrislawn mentioned, you can see this process illustrated in the FMT tutorial.
Hope this helps!
Thank you.
I’m working through it and I’ve come to the part where I would need to create a summary of the merged table.
I have two different metadata files (mapping).
If I were to just combine these files would it become an issue that I have two samples with the same barcode?
When I ran it through the taxonomy portion they went through and did not give me an error, I just want to be 100% sure.
You’re fine — the barcodes are only used for demultiplexing. You could drop that column if you like (you might want to, since it will show up in downstream steps like those in alpha-group-significance as Metadata Columns, which probably won’t have any useful meaning in the context of your analysis). Keep us posted! 
This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.