How does your sequencing center communicate the relationship between samples, flowcells/lanes (or other sequencer setups), and barcodes. Given the many-to-many schemes that people cook up this seems like a huge source of error.
Is there a data format and/or LIMS that a technician can easily use for this purpose?
We use a custom LIMS, but we model Sources, Samples, Libraries (which can be derived from other libraries), Runs (collections of lanes and libraries on a flowcell), and Files. The LIMS also has a concept of a Study, which is a collection of Samples (and, by association, files). There is a lot of detail in the system, but that is what we ended up with. Tag/Value pairs are used for annotation and metadata that is not incorporated into the model elsewhere.
You could also look at ENA and SRA for their concepts of metadata.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
Something that drives me nuts is a lack of documentation coming from collaborators. I know it's not meta data, but at least you could tell me: did you trim the reads? what is the platform? who should I send correspondence to? A hard drive containing unknown fastqs is a daily nightmare.