GPL-387 As a user I need to be able to create an ONT run

[stevieing]

User story As a user I need to be able to create an ONT run

Who the primary contacts are for this story Harry/Steve

Acceptance criteria To be considered successful the solution must allow:

Run required fields:

• Instrument name
• Experiment name
• State

A run has many flowcells. Flowcell required fields:

• Flowcell ID
• Flowcell position
• Kit selection
• Basecall model (High accuracy / Fast / No basecalling)

A flowcell has a library Library required fields: ...

• Successfully created a pacbio run in the database
• can GET run/runs

Additional context What events are created? Any automation with ONT API?

Breakdown of tasks

• Generate architecture of active record models.
• Create a factory for generating runs.
• Create GraphQL endpoints.
  • Retrieve libraries with filter to exclude those already associated with a flowcell.
  • Mutation to create runs.
• Update message specifications for ONT runs.
• Message over Rabbit when runs are created as part of the above mutation.

[harrietc52]

On run creation

• set experiment name
• create flowcells
• each flowcell have a uuid, position
• each flowcell is associated with a library
• send a message to the MLWH

Unknowns

• store instrument name?
• store state?

MLWH

• Send message to MLWH
• Do NPG require anything for multiplex? This will affect what sample meta data we pull from SS and is required on library/run creation

[Chris-Friend]

GPL-387 is the feature branch for this story. The storyy will be complete on merging this feature branch into develop

Tasks breakdown:

• @Chris-Friend: Add Ont::Run and Ont::Flowcell active models. The run should store the instrument name (fixed and from config), the experiment name (id), state (see Saphyr run, e.g. 'pending'), and should have many flowcells. The flowcell a position, a uuid and must have a single library. The library relationship to a flowcell should be optional
• @Chris-Friend: Add run and flowcells factories. This should take a list of library identifiers (names or ids) and generate a single run with many flowcells containing a single library
• Add new GraphQL endpoints. One to create an ont run. Update the existing libraries endpoint to allow returning only libraries that have not been assigned to a flowcell
• Send flowcells message(s) to the MLWH. This will require adding a new ont flowcells message config in the pipelines.yml, and should be called in the GraphQL controller after successful creation of the run. Generated message should be tested

[stevieing]

The message to the warehouse should contain the following fields. For each flowcell there will be a separate entry for each request in a library:

• id_flowcell_lims: flowcell.id,
• updated_at: timestamp,
• sample_uuid: sample.external_id,
• study_uuid: request.external_study_id,
• experiment_name: run.experiment_name,
• instrument_name: hardcoded e.g 'GXB12345',
• instrument_slot: flowcell.position,
• pipeline_id_lims: library_preparation_type (don't know where this is coming from at the moment),

for tagging purposes (source to be determined):

• tag_identifier
• tag_sequence
• tag_set_id_lims
• tag_set_name

If there is a second tag:

• tag2_set_id_lims
• tag2_set_name
• tag2_identifier
• tag2_sequence