FAIR Data Registry

FAIR Data Registry #

The FAIR data pipeline requires to maintain a data registry as we have the following objectives:

  • We need to identify and set up a system which will let everything be managed, version controlled, etc, etc.
  • We will need to co-ordinate people to curate incoming data, generating version controlled data resources in formats required by the modellers.
  • We will need to co-ordinate people to curate incoming data, generating version controlled data resources in formats required by the statisticians.
  • We will need to co-ordinate people to curate papers and reports, so these are available for critical review.
  • We will need to co-ordinate statisticians to analyse the data curated under point (3), producing version controlled statistical estimates of key variables relevant to the parameterisation of the models, to a prescribed standard.
  • We will need to co-ordinate reviewers to scrape estimates of key variables relevant to the parameterisation of the models from the literature, to a prescribed standard.
  • With the modellers, we will need to identify appropriate functional forms to map these version controlled statistical estimates to the parametric form required by the models, and incorporate these into the workflows.
  • We need a transparent mechanism to either allow modellers to choose parameter values from a portfolio of version controlled values, or to integrate these into a single ensemble estimate (preferably taking account of the credibility and relevance of each source).