Y26-127 - As a manager in short read in charge of billing, I want the billing report to correctly handle Bioscan samples.

[dn10-sanger]

User Story
As a manager in Short Read in charge of billing, I want the billing report to correctly handle BioScan samples, so that lane proportions and totals are accurate when ToL begin using the billing report.

Acceptance Criteria
TBD pending stakeholder input

Who the primary contacts are for this work
Tom W
Ivy T
Laura H
Rich R
Dikshita N

Context
The billing report joins stock_resource to retrieve stock plate barcodes. For Bioscan samples, a single sample can be associated with more than one stock plate (e.g. both a LILYS plate and a Lysate plate are registered as stock plates in stock_resource)- see sanger/sequencescape#4706 for where it was implemented. This causes the sample to appear multiple times in the sample_lanes CTE, which inflates the COUNT(*) used to calculate lane proportions resulting in incorrect total values for Bioscan runs.

This is a known limitation and was flagged during development of #51. ToL are not currently using the billing report so this is not urgent, but it must be resolved before they do.

Suggested two possible approaches

1. Show one row per sample and choose the stock plate barcode most useful for billing.
2. Show one row per sample and concatenate multiple stock plate barcodes into one field.

Known Bioscan pipeline routes:

1. Extraction performed in SeqOps :
   Service user provides the LILYS-96 plate to SeqOps, and SeqOps perform the extraction and produce the Lysate
   plate -> two stock plates expected since Y25-157 - ToL BIOSCAN - Register LBSN-96 Lysate plates as stock resources in MLWH sequencescape#4706. Only one (LILYS) expected before that, unless data was back-filled.
2. Extraction performed in programmes :
   Faculty perform the extraction themselves and provide the Lysate plate to SeqOps - > one stock plate expected (Lysate)

Research Task

1. Inspect Bioscan sample data in stock_resource to understand how often samples have one vs multiple stock plates.
2. Review examples covering the two known pipeline routes.
3. Confirm with stakeholders whether to pick one stock plate or concatenate multiple barcodes.
4. Use findings to define the implementation approach and write acceptance criteria for the fix.

Reference
sanger/ sample-tracking #51

[KatyTaylor]

Hi @TWJW-SANGER , @dn10-sanger and I are wondering if this needs to be prioritised. Do you have any info on priority other than what it says in the story ("ToL are not currently using the billing report so this is not urgent, but it must be resolved before they do.")? I couldn't find any relevant comments in email threads.

Thanks

[TWJW-SANGER]

@KatyTaylor Hi. This came from a conversion I had with Rich R about this on the way to another meeting.
ToL seem to have some other billing arrangement, possibly because they make heavy use of PacBio Long Read.
I am not aware of any plans for this to change soon. But if any of us catch up with Rich R, it would be good to check.