MiniStd Call 2020-01

Agenda

DataRep decision on organism field
DataRep is now the owner MiAIRR Set 6 fields
Object definition for receptor and cell (see Christian's comment of 2019-12-24 on #273)
List of To-Does for MiAIRR v2 (#305)

Should fields be non-nullable based on the availability of the information to the primary data depository (current situation) or the necessity of the information for meaningful interpretation? Note that the current situation can make it hard for third-party annotators (#310).
- A few things that non-nullable status could indicate:
  - Criticality to MiAIRR as a Standard: Fields which one MUST always have, as decided by the AIRR Community.
  - Field one always is expected to have: Not necessarily critical to MiAIRR, but hard to understand how one could do a study and not have it...
- Noted that many of the non-nullable fields are controlled vocabularies with NULL like options such as library_generation_method:other and physical_linkage: none. Perhaps for non-nullable fields this should be the norm. We should consider carefully those fields that have limited possible values (booleans, controlled vocabularies lacking NULL-like terms) and ensure that if they do not exist, we really want that data be not AIRR-compliant.
Should we switch notes from Google Docs to Github?
Review CEDAR Templates

Approved Human Population Genetics XT (#264, #265)
Approved moving/introducing the fields ancestry_population, country_birth and collection_country to/in an Extension.
As ethnicity and race have neither a consistent scientific concept nor globally applicable ontologies, they are removed from MiAIRR and its extensions. Note that annotators who wish to provide this information can still do so using these keywords as optional free text fields.
The integration of extensions into the schema still needs to be discussed with DataRep. Therefore a first draft has now been commited (#318).

DataRep deferred the decision on whether to rename organism to species, will bring it up again in MiniStd Call 2020-02.
DataRep has acknowledged that they are now the owner of the fields in Set 6 (see minutes of MiniStd Call 2019-11 and MiniStd Call 2019-12)
We are now collecting things that need to be included for MiAIRR v2 in #305. In most cases the things will/should also have an entry of their own on the issue tracker, in which case these should be labeled with the AIRRv2.0 and the MiAIRR tag in addition.
cell and receptor objects (#273, #211, #206): There is now an emerging consensus based on Christian's comment of 2019-12-24 on #273. This has been approved by DataRep, Sri is now working on a schema definition. Note that pair_id never made it into an official release, thus it is simple to deprecate it.

Revisit MiAIRR non-nullable fields (#310): Currently non-nullable status (aka required) is based on the near-certain availability of the information to the primary data depository. However, it turns out that this makes it hard for third-party annotators, therefore it has been proposed to revisit these fields based on the criterium whether the information is strictly required for meaningful interpretation of the annotated data.
John will soon make the CEDAR AIRR templates publicly available and asks for comments (link to CEDAR Templates). Note that these templates are identical to the information on the actual CEDAR submission site, it is just accessible without requiring a login. In case you would like to comment on this, please get in contact with John until Thu, 2020-01-23.
Discussed whether it would be worthwhile to put the agendas and minutes on Github instead of GDocs. This would resolve some of the overhead that the current workflow produces. Brian comments that ComRepo has experimented with this, but not adapted a Github workflow as copyediting can be an issue as documents will be public. Will discuss again in the next call.