Core Concepts
The rest of this documentation is task-oriented: it tells you which screen to open and which button to press. This section is the layer underneath — what the objects are called, why the workflow is shaped the way it is, and what SeqDesk deliberately refuses to do.
If a page elsewhere in the docs uses “Sequencing Order” and “Sequencing Run” in the same sentence and you are not sure they are different things, this is the section that fixes that. Start with How SeqDesk Works.
The one-paragraph version
A researcher raises a Sequencing Order: a request carrying a list of
Samples. The facility sequences those samples on a Sequencing Run and
attaches the resulting FASTQ files as Reads. Samples — possibly from
several orders — are grouped into a Study, which is the scientific unit and
the thing that carries MIxS metadata. Nextflow pipelines run over an order
or over a study. Finished studies are submitted to the European Nucleotide
Archive. Two roles exist: RESEARCHER and FACILITY_ADMIN. Everything runs on
your own hardware.
Where to start
| If you are | Read |
|---|---|
| New to SeqDesk entirely | How SeqDesk Works, then the Getting Started quickstart for your role |
| Evaluating whether SeqDesk fits your facility | What SeqDesk doesn’t do (yet) first — it is faster to rule out than to rule in |
| About to onboard real researchers | Best Practices |
| Preparing a training session or a demo | Demo Mode |
Three ideas that explain most surprises
Newcomers hit the same three walls. Knowing about them in advance saves an hour of confusion each.
Submitting an order freezes its samples. While an order is DRAFT you can
add, edit, remove and bulk-import samples freely. The moment it becomes
SUBMITTED, the samples API refuses structural changes with
Cannot modify samples after order submission. Only facility-only fields stay
editable, and only for admins. Get the sample table right before you submit.
Researchers see results only after an admin publishes them. Sequencing files are invisible to the requesting researcher until an admin presses Make downloadable to user on the order, and even then only cleaned reads are handed over. Pipeline runs are invisible until an admin marks one as Final for that pipeline and target. Nothing leaks to the researcher by accident — which also means nothing appears by accident. If a researcher reports “I cannot see my data”, this is almost always why.
A sample belongs to at most one study. Assigning a sample to a second study moves it out of the first. Studies are not tags.
In this section
Neighbouring sections
Once the vocabulary makes sense, the feature sections pick up where this one stops: Sequencing Orders & Samples and Studies cover the two halves of the data model, Sequencing Files covers how FASTQ files get attached, Pipelines & Analysis covers Nextflow execution, and ENA Submission covers publication. For field-level schema detail, go to Reference → Data Model.