AllenNeuralDynamics / AllenNeuralDynamics/biodata-schema

Curation.json

Abierto
#38 0 comentarios 0 reacciones 1 asignado Asignado a @dbirman Ver en GitHub
Lenguaje dominante
Python
Estrellas
0
Forks
0
Métricas de merge de PR
Sin PR fusionados en 30 d

Descripción

Users want to be able to "annotate", "curate", and QC data that is an "element" of a data asset but not the entire asset. For example:

- after histology, ephys users want to annotate the real probe track location somewhere
- after spike sorting users need to be able to merge/split units, curating the spike sorting
- in ophys data individual ROIs have per-roi quality control done
- during exaspim reconstruction neurons need to be annotated and there can be thousands of neurons

We currently enable this with the `CurationMetric` but we are finding this unwieldy due to issues with size and supporting the variety of use cases (see [here](https://github.com/AllenNeuralDynamics/aind-data-schema/pull/1608), for example where I'm grappling with how to effectively compress the curations to deal with exaspim properly).

I think that we need to consider making annotation/curation a core concept in the schema, which would make it easier to add a variety of classes supporting the different use cases we're running into. This would also allow us to enforce standardization on some kinds of annotations, for example ephys probe track annotations should ideally be in an atlas coordinate system! It would also make it more obvious to users where we expect them to put these kinds of things which are currently hidden in the QC.

Guía de contribución

Abrir la guía de contribución

Evaluación

Este issue todavía no se ha evaluado.

Recibe los nuevos issues en tu correo

Un resumen breve de issues de GitHub para principiantes.