Skip to Content
Sequencing Orders & SamplesOverview

Sequencing Orders & Samples

A Sequencing Order is the record of one sequencing request: who asked for it, which samples are involved, what should be run on them, and — once the facility is done — which files came back. Everything else in SeqDesk hangs off it. Samples cannot exist without an order, sequencing runs are created inside an order, and read files are linked to samples that belong to an order.

Naming. The database model is Order and its identifiers are orderId / orderNumber (ORD-YYYYMMDD-XXXX), but the interface calls it a Sequencing Order. A SequencingRun is a different thing — the physical run on an instrument. One Sequencing Order contains zero or more Sequencing Runs. See Sequencing Runs.

The mental model

Read this once and the rest of the section follows:

Sequencing Order (ORD-20260415-0001, status DRAFT → SUBMITTED → COMPLETED) ├── Sample (S-1755262041412-K7QX2, one per tube) │ ├── Read (file1 / file2 — the FASTQ pair linked to this sample) │ ├── Assembly / Bin (produced later by pipelines) │ └── Study (optional, at most one — the analysis / publication grouping) └── SequencingRun (the instrument run; maps barcodes → samples)

Three facts that explain most of SeqDesk’s behaviour:

  • A sample belongs to exactly one order, forever. Sample.orderId is required and is never reassigned. If a tube was ordered under the wrong order, the sample has to be recreated.
  • A sample belongs to at most one study. Sample.studyId is nullable. Studies are how you regroup samples across orders for analysis and ENA submission — see Studies.
  • Reads are attached to samples, not to orders or runs. A SequencingRun records the barcode plan; the actual FASTQ paths live on Read records pointing at samples.

Who does what

SeqDesk has two roles that matter here, and the split is strict.

Researcher (RESEARCHER)Facility admin (FACILITY_ADMIN)
Create a sequencing orderYesYes
Edit sample data (organism, titles, MIxS, custom fields)Only while DRAFTOnly while DRAFT
Edit order-level metadataUntil COMPLETEDAlways
Edit admin-only fields (Internal QC Status, Sample QC Result, …)Never — they are invisibleAlways
Create sequencing runs and run plansNeverOnce SUBMITTED
Discover, associate and upload read filesNeverOnce SUBMITTED
See the read files at allOnly after the facility publishes themAlways
Delete a submitted orderNeverOnly if the admin setting allows it

That last row is the one people trip over: until a facility admin presses Make downloadable to user, a researcher’s order page shows no read files at all, even though the files are linked in the database. See Sequencing Order Lifecycle.

Where to start