Can QIIME2 split library or combine library

[sdpapet]

Hello, I did several MiSeq amplicon seuqencing (16S/ITS). The sequencing center general three files, Forward fastq, Revsere Fastq and barcode fastq. which are prefect using QIIME2 tutorial (Tutorials — QIIME 2 2021.11.0 documentation)

I am not having trouble in building ASV table so far. I am wondering if QIIME2 supports to split a library into individually libraries at the very beginning. All of the QIIME 2 tutorials are using the total big file. Bascially, until I generate ASV table, I cannot filter or remove samples.

In my case, I have one library. There are half of samples are mine, the other half of samples are from the other lab. Is it possible for me to split individual fastq files and combine mine samples together. So, I can give the rest samples to the other labs? One of my friend told me QIIME1 can do this.

Also, is it possible for QIIME2 to start workflow (e.g. DADA) with individual already spitted libraries?

[timanix (Timur Yergaliyev)]

Hello!
It is recommended to process big files to create ASV table and rep-seqs file since Dada2 needs as much sequences from the same run as possible. In other words, splitting 2 datasets before Dada2 can lead to biases in the output.
So, one option will be to demux files first, remove primers and then denoise all samples together with in Dada2, and only then filter output tables by metadata file on different datasets. You can also filter rep-seqs files for each dataset based on resulted feature tables.
If second team is not going to use Dada2 for denoising, or wants to have raw data, you also can export all the samples from demultiplexed qza file and separate them manually as a bunch of fastq.gz files.