際際滷

際際滷Share a Scribd company logo

The Biological Data Sustainability Paradox: A Time to Think Differently

Download as PPTX, PDF
Philip Bourne
Philip Bourne

Presentation to the ISCB Fellows on aspects of data that the society should be concerned about.

1 of 14
The Biological Data Sustainability Paradox: A Time To Think Differently Terence Johnson & Philip E. Bourne School of Data Science & Dept of Biomedical Engineering University of Virginia https://arxiv.org/abs/2311.05668 ISCB Fellows Meeting July 2024 1
Global Data Growth Over 10 Years 130 Zettabytes ISCB Fellows Meeting July 2024 2
The Paradox With a fixed funding base the maintenance of biomedical data takes money away from innovative research, however Those data are increasingly important in achieving innovation. ISCB Fellows Meeting July 2024 3
Today - Biological Data Sustainability (BDS) 1.0 Funding of biological data is mostly with public money Public private partnership is limited Scientific foundation models are limited The system is fragile think USA November 2024 Some gains in efficiency are possible through aggregation/centralization Politics, cultures and funding models work against aggregation/centralization ISCB Fellows Meeting July 2024 4
Funders Want To Do Something https://globalbiodata.org/what-we-do/global-core-biodata-resources/ ISCB Fellows Meeting July 2024 5
Tomorrow Biological Data Sustainability (BDS) 2.0 A data economy Not for one moment do we think the community of funders and data providers and users will find this model palatable The intent is to invoke discussion amongst a broader group of innovative thinkers ISCB Fellows Meeting July 2024 6
Biological Data Sustainability (BDS) 2.0 Cap and Trade Model Premise Data have value in the marketplace and can be traded That value is in the form of currency - cash or data credits A global broker perhaps a coalition of funders distributes credits Services have value in the marketplace The private sector is an active not passive player in the model The era of entitlement by all stakeholders researchers, data resources, funders is over whatever the model, it is driven by the economics of the marketplace ISCB Fellows Meeting July 2024 7
BDS 2.0 Mechanics Funders act collectively to support a global broker Every researcher get some free data credits Data upload to a repository or data download expends credits Data download is paid in credits by the downloader Data depositor received credits upon that download minus an operational fee Parties can solicit data credits by performing services Data credits can be exchanged for cash for which the broker gets a fee The broker issues or removes credits in the system to maintain stability ISCB Fellows Meeting July 2024 8
BDS 2.0 Mechanics ISCB Fellows Meeting July 2024 9
BSD 2.0 Implications Controls the amount of data in the system A single currency for evaluating the system A depositor dividend for producing widely used data Disincentivizes depositing low quality data as its costs credits but will not gain credits Creates a distributed service economy supporting curation Private sector actors who do not deposit pay cash for credits or perform services Access to the system is tied to data quality ISCB Fellows Meeting July 2024 10
BDS 2.0 Advantages Multiple brokers a problem but maybe it means its working Anyone e.g. in US parlance R2s, NBCUs, community colleges can participate Creates a bias for producing highly accessed datasets Potential for higher quality curation Red is an advantage and disadvantage ISCB Fellows Meeting July 2024 11
BDS 2.0 Disadvantages Its an all or nothing approach and hard to launch Predatory data purveyors Multiple brokers multiple non-consistent markets Creates a bias for producing highly accessed datasets Potential for lower quality curation Red is an advantage and disadvantage ISCB Fellows Meeting July 2024 12
Is BDS 2.0 More Sustainable than BDS 1.0? 2.0 leverages the entire community of users A distributed service economy will raise overall awareness concerning data quality ISCB Fellows Meeting July 2024 13
Next Steps There is a GBC Funders meeting in September at which I will present and the model discussed. It would be helpful to have a view from the ISCB community (positive or negative) to add to the discussion. Publish a paper based on the arXiv version and all input received. We (well mainly Terry Johnson) has developed a mathematical model to simulate the activities of the credits system. We will report back. ISCB Fellows Meeting July 2024 14

More Related Content

The Biological Data Sustainability Paradox: A Time to Think Differently

  • 1. The Biological Data Sustainability Paradox: A Time To Think Differently Terence Johnson & Philip E. Bourne School of Data Science & Dept of Biomedical Engineering University of Virginia https://arxiv.org/abs/2311.05668 ISCB Fellows Meeting July 2024 1
  • 2. Global Data Growth Over 10 Years 130 Zettabytes ISCB Fellows Meeting July 2024 2
  • 3. The Paradox With a fixed funding base the maintenance of biomedical data takes money away from innovative research, however Those data are increasingly important in achieving innovation. ISCB Fellows Meeting July 2024 3
  • 4. Today - Biological Data Sustainability (BDS) 1.0 Funding of biological data is mostly with public money Public private partnership is limited Scientific foundation models are limited The system is fragile think USA November 2024 Some gains in efficiency are possible through aggregation/centralization Politics, cultures and funding models work against aggregation/centralization ISCB Fellows Meeting July 2024 4
  • 5. Funders Want To Do Something https://globalbiodata.org/what-we-do/global-core-biodata-resources/ ISCB Fellows Meeting July 2024 5
  • 6. Tomorrow Biological Data Sustainability (BDS) 2.0 A data economy Not for one moment do we think the community of funders and data providers and users will find this model palatable The intent is to invoke discussion amongst a broader group of innovative thinkers ISCB Fellows Meeting July 2024 6
  • 7. Biological Data Sustainability (BDS) 2.0 Cap and Trade Model Premise Data have value in the marketplace and can be traded That value is in the form of currency - cash or data credits A global broker perhaps a coalition of funders distributes credits Services have value in the marketplace The private sector is an active not passive player in the model The era of entitlement by all stakeholders researchers, data resources, funders is over whatever the model, it is driven by the economics of the marketplace ISCB Fellows Meeting July 2024 7
  • 8. BDS 2.0 Mechanics Funders act collectively to support a global broker Every researcher get some free data credits Data upload to a repository or data download expends credits Data download is paid in credits by the downloader Data depositor received credits upon that download minus an operational fee Parties can solicit data credits by performing services Data credits can be exchanged for cash for which the broker gets a fee The broker issues or removes credits in the system to maintain stability ISCB Fellows Meeting July 2024 8
  • 9. BDS 2.0 Mechanics ISCB Fellows Meeting July 2024 9
  • 10. BSD 2.0 Implications Controls the amount of data in the system A single currency for evaluating the system A depositor dividend for producing widely used data Disincentivizes depositing low quality data as its costs credits but will not gain credits Creates a distributed service economy supporting curation Private sector actors who do not deposit pay cash for credits or perform services Access to the system is tied to data quality ISCB Fellows Meeting July 2024 10
  • 11. BDS 2.0 Advantages Multiple brokers a problem but maybe it means its working Anyone e.g. in US parlance R2s, NBCUs, community colleges can participate Creates a bias for producing highly accessed datasets Potential for higher quality curation Red is an advantage and disadvantage ISCB Fellows Meeting July 2024 11
  • 12. BDS 2.0 Disadvantages Its an all or nothing approach and hard to launch Predatory data purveyors Multiple brokers multiple non-consistent markets Creates a bias for producing highly accessed datasets Potential for lower quality curation Red is an advantage and disadvantage ISCB Fellows Meeting July 2024 12
  • 13. Is BDS 2.0 More Sustainable than BDS 1.0? 2.0 leverages the entire community of users A distributed service economy will raise overall awareness concerning data quality ISCB Fellows Meeting July 2024 13
  • 14. Next Steps There is a GBC Funders meeting in September at which I will present and the model discussed. It would be helpful to have a view from the ISCB community (positive or negative) to add to the discussion. Publish a paper based on the arXiv version and all input received. We (well mainly Terry Johnson) has developed a mathematical model to simulate the activities of the credits system. We will report back. ISCB Fellows Meeting July 2024 14