CaTissue has a facility for automatic labelling (and barcoding) of specimens. We have had this turned off because we don't want CaTissue to generate barcodes, as the aliquot storage tubes are all barcoded on manufacture. However, it turns out we can have auto labelling, without auto barcodes. This would speed up data entry for projects (other than BRICCS, because of the API client doing this automatically anyway).

So... we need to test our existing labelling scheme for compliance with the token system available in CaTissue.

Using the DEV deployment...

1. GENVASC

* Primary samples: 

  * prefix "s" for the whole blood which generates the serum sample (collected in serum separator vacutainer)
  * prefix "e" for the whole blood which generates the plasma sample (collected in edta vacutainer)
  * Collection Protocol ID (e.g. "G00007") 

  * Thus: "sG00007" and "eG00007"

* Derived samples 

  * Primary sample label
  * Alphabetical ordinal ("a", "b", "c", etc)

  * Thus: "sG00007a", "eG00007a" and "eG00007b"

* Aliquots

  * Manually entered 2D barcodes.


Problems:
  1. The token system available doesn't differentiate by container, only by specimen type.
  2. Ordinals are only available as numbers, not letters.
  3. DNA is derived - just check what happens, will the DNA be auto-labelled? That would probably be OK.

2. BIOSTAT

No auto-labelling required - all specimen arrive in 2d barcode tubes.

3. Prestige

* Only one primary sample:

  * Collection Protocol ID

* Derived samples

  * Primary sample label
  * Suffix of "-" followed by the sample type abbreviation, using P for plasma and B for Buffy coat.

* Aliquots

  * Manually entered 2D barcodes.

Problems:
  1. Derived samples use different abbreviations to the BRICCS and GENVASC protocols, and abbreviations.xml is a site-wide configuration file.
  2. The abbreviations a bit odd - "B" for Buffy Coat, and P for "edta plasma". But it is the only plasma in this CP.

4. DREAM

* Primary samples (two visits, two samples per visit)

  * Collection Protocol ID
  * Alphabetical ordinal ("a", "b", "c", etc)

* Derived samples

  * IT SHOULD BE: Parent specimen ID with a suffix for specimen type ("p" for plasma and "s" for serum).
  * But there's a problem. The first visit specimens are being labelled with the CP ID and the suffix, which means the second visit specimen will have the same labels.

Problems:
  1. Ordinals are only available as numbers, not letters.
  2. The labelling scheme for the derived samples needs fixing.


5. MVO

* Primary samples

  * Collection Protocol ID
  * Suffix of "-" followed by an ordinal relating to the visit event

Problems
  1. Ordinal based on visit event isn't currently possible. If third visit is missed, next sample should be "4" but would get a "3". Could we add a token to include the 

6. BRICCS-HDL - not yet recruited. Real soon now.

* Primary sample (additional EDTA): BSa number as the specimen is obtained during BRICCS recruitment

* Derived sample: prefix "e" for edta plasma, followed by the Collection Protocol ID.