Upgrade to Pro — share decks privately, control downloads, hide ads and more …

HCA General Meeting 2022: Cell Annotation Platform

Evan Biederstedt
June 27, 2022
320

HCA General Meeting 2022: Cell Annotation Platform

Talk at the HCA General Meeting 2022. Cell Annotation Platform updates.

Evan Biederstedt

June 27, 2022
Tweet

Transcript

  1. Cell Annotation Platform Evan Biederstedt Department of Biomedical Informatics De

    fi ning Cell Types and States for the Human Cell Atlas and Beyond
  2. NK Cells Cytokines Monocytes Problem •Individual research groups end up

    annotating (potentially millions of) cells manually, which results in cells with inconsistent terms and labelings between groups. •This approach cannot scale. We need a solution for creating comprehensive references with a standardized nomenclature for all species. •There's no medium for researchers to compare annotations across studies, potentially resolving con fl icting results. •There’s no central location to access annotations used in publications. Motivation
  3. Denis Ilguzin Maxim Svetlakov Levon Ghukasyan Michael Loktionov Sultan Arapov

    Mary Futey Nick Akhmetov Anush Boyakhchyan Tigran Markosjan Konstantin Boyandin Uğur Bayindir David Osumi-Sutherland Pavel Istomin Dennis Bolgov Felix Fischer Mo Lotfollahi David Fischer Evan Biederstedt
  4. Cell Annotation Platform (CAP) • Community-driven platform to create, explore,

    and store annotations • Infrastructure to accumulate, share, and analyze annotation terms with associated molecular signatures to interpret cellular identities • Encourage researchers to converge upon consensus nomenclature • 
 •
  5. Cell Annotation Platform (CAP) • Standardized formats, APIs, and citable

    accession IDs for published cell annotations • Enable automated cell annotation service of cell types and cell states • Reference for cell identities based on molecular signatures across species
  6. Cell Annotation Platform (CAP) Main Components • Data Repository •

    Annotation Upload and Publication • Annotation UI • “CellCards” Reference Summaries •
  7. MVP User Workflow 
 1. Sign in & User Profile

    2. Upload already annotated data 3. Collaboratively edit and save 4. Publish version (with DOI) 5. Downloadable results 6. Browse / Search Current Release https://celltype.info/
  8. MVP User Workflow 
 1. Sign in & User Profile

    2. Upload already annotated data 3. Collaboratively edit and save 4. Publish version (with DOI) 5. Downloadable results 6. Browse / Search Current Release https://celltype.info/
  9. CAP organization • Workspace: Collaborative “repo” for researchers to organize

    annotations
 • Publication: Version • Datasets: Cell annotations with molecular data • Cell Label: Term associated with a cell or molecular subpopulation.
  10. • Collections of datasets, typically corresponding to a scienti fi

    c journal article • Timestamped
 • DOIs for citations in journals
 • Versioning • Downloaded annotations in standardized formats 
 Publications
  11. Workspace • Collaborative space to edit collections of annotations &

    other relevant metadata • Advanced user form • Allow user to “hide” irrelevant metadata within dataset 

  12. • Specify which annotations & which metadata fi elds are

    relevant
 • Allow user to “hide” irrelevant metadata within dataset 
 Workspace
  13. 
 • Autocomplete recommendations (with synonyms and related terms) from

    EMBL-EBI ontologies
 • “Nudges” to encourage consensus and standardization (if possible) but no requirements 
 Workspace
  14. • Users roles for collaborative work on annotations
 • User

    roles: • viewer (read-only) • editor (write access) • owner (administrative) • 
 • Collaborations
  15. Cell Synonyms & Categories • Synonyms
 • Categories 
 e.g.

    “CD8+ T cell” is a subset of “T Lymphocyte” • Relationships between annotations • 
 •
  16. Cell Synonyms & Categories • Synonyms
 • Categories 
 e.g.

    “CD8+ T cell” is a subset of “T Lymphocyte” • Relationships between annotations • Why? Discover patterns of how the community is naming these entities.
  17. Cell Annotation Platform (CAP) Main Components • Data Repository •

    Annotation Upload and Publication • Annotation UI • “CellCards” Reference Summaries •
  18. CAP Timeline: User Work fl ow User registers/signs into CAP,

    and uploads a dataset which has been annotated. AnnData & Seurat + Cell Annotations + Current Release (June 2022)
  19. CAP Timeline: User Work fl ow AnnData & Seurat +

    ___________ + Upcoming Release (Autumn 2022) User registers/signs into CAP, and uploads a dataset which has NOT been annotated.
  20. CAP Timeline: User Work fl ow Automated annotation suggestions in

    real-time AnnData & Seurat + ___________ + Uploads a dataset which has NOT been annotated.
  21. • sfaira.data: formatted AnnData objects for model input, control meta

    data in ontologies. • sfaira.models: supervised models of cell types • atlas-based label transfer via query- reference projection (scArches, CellTypist, Azimuth, Human Lung Cell Atlas HLCA, …) e.g.: deploy models fi t per organ to annotated organ of query sample Fischer, Dony, et. al., “sfaira”, 2020 Predictive modeling of cell types
  22. Next Steps • Autumn 2022: Annotation UI with real-time cell

    type suggestions
 • Beginning 2023 • Summaries via “CellCards” pages • Advanced UI + analysis features

  23. User Feedback Request • Prioritize Future Features • Annotation Feedback?

    • Community Ratings? • Evidence? • Contrast/Compare Annotation A vs B? • Speci fi c UI Requests?
 • Demos & User Feedback Surveys
 Talk to us!