NExtSEEK
All docs pages

Sample types

Every record in NExtSEEK is a sample, and every sample has a sample type. The type is a template: it says which attributes the sample carries, which types it comes from and which types come from it.

What a sample type is

A sample type is a named schema. A mouse and a tissue sample are different types because they carry different fields. When you upload samples, you pick the type, and the upload sheet has the columns that type defines.

Each type has:

  • a code, the short prefix of every UID of that type
  • a name and a description
  • a clade (see below)
  • attributes, some required and some optional
  • parent types (derives from) and child types
  • tags, the extra words people search by

The code

The code is the prefix of the sample's UID. A mouse sample has a UID that starts with MUS, a sequencing data file with D.SEQ, and a gene expression analysis with A.GEX.

Code Name Clade
NHP Non Human Primate Source
TIS Tissue Sample Processed
D.SEQ Sequencing Data Raw
A.GEX Gene Expression Analysis Analyzed

NExtSEEK makes the UID when you upload a new sample. You do not make them up. See Uploading.

The four clades

Sample types are grouped into four clades. A clade is a stage in the life of the data, from the thing you start with to the result you publish. The data model page shows how the clades connect.

Clade What it holds Code style
Source Organisms and starting materials: mice, cell lines, patients, chemicals Short codes: MUS, CEL, PAT, CHM
Processed Physical products made from a source: tissue, DNA, RNA Short codes: TIS, DNA, RNA
Raw Instrument output, one data-file type per kind of measurement D. prefix: D.SEQ, D.FLOW, D.IMG
Analyzed Results of analyzing raw data, and models A. prefix: A.GEX, A.TITR. Models use M. or MDL

Required and optional attributes

Every attribute has a name, a type (text, date and so on) and a definition. Some are required: a sample cannot be uploaded without them.

  • Every type requires a UID. Almost every type also requires a Scientist, the lab member responsible, and Source and Processed types require a Name.
  • Types in the Raw and Analyzed clades, and most Processed types, also require a Parent, the sample they came from.
  • Raw and Analyzed data-file types also require the file, a link to it and, for nearly all of them, a checksum.
  • Everything else is optional. Fill in as much as you know: optional fields are what make samples findable later.

The detail page of each type lists its attributes, marks the required ones and shows the definition of each.

Parents and children

A type lists the types it derives from and the types that derive from it. A patient visit (PAV), for example, derives from an NHP, a patient or a mouse, and tissue (TIS) derives from a patient visit. An assay is what turns parents into children: see Assays.

Tags

Tags are alternative names for a type ("macaque", "rhesus macaque" and "primate model" for NHP). Search and Nessie use them to match what you type to the right type.

Browse every sample type

The live catalog is always current. It lists every type grouped by clade, with a filter box.

Open the Sample Types catalog (you need to be signed in).

The Sample Types catalog page: a search box above a table of sample types grouped by clade, starting with the Source clade. Each row shows the code, name, number of attributes, number of samples and a description.
The catalog, grouped by clade.

Click a code to open the detail page. It shows the definition, what the type derives from, its children, the assays that produce and consume it, its tags and every attribute. From there you can download the upload template for that type or search its samples.

The detail page of the NHP sample type: its description, then rows for derives from, children, produced by, consumed by and tags, then a table of attributes with required fields marked.
One type's detail page.

Need a new type or attribute?

Contact the data team from the docs menu. Administrators can edit attributes on the Admin pages.