How to archive sequencing runs with multiple experiments

Hi all,

I am looking for some input on best practices when it comes to uploading raw data from sequencing runs to repositories such as SRA, ENA or QIITA when we have multiple experiments on one chip.

It seems that these archives allow multiple runs to belong to a single project/experiment but not vice versa. I would simply split up the samples and upload separately, but they were processed together and shared some negative controls which I need to use and reference in two publications. Duplicating those controls does not seem like an ideal solution.

A method that allows the entire output of the sequencing run to be stored and identified as a single entity would also be ideal as I used the whole dataset to train my error model in DADA2.

Any experiences with this problem or advice on an archive that provides a suitable way to upload this data would be much appreciated.


Hi @Boyd,

Welcome to the :qiime2: forum!

Most projects Ive worked on have kept separate experiments separate. (We might sequencing 600 samples together but my project might have only used 200.) This is actually a really common phenomeon for Qiita: there’s even a data upload section that lets you use the same fastq files for multiple studies. (Although Qiita also uses deblur because its data independent.)

In your case, i think i would upload the positive and negative controls with each study. However, you could also contact EBI to see if they have recommendations. They’re help account has been fairly responsive and helpful when I’ve worked with them, and they may have stronger preferences about what they want. (And if you do and want to report back, we’d love to know.)



Thanks so much for the advice Justine.

It sounds like Qiita will be able to do exactly what I want, but I will also get in touch with EBI and share any advice they provide.

1 Like