Surveys of IEDB Experimental Types - SciKnowEngine/SciKnowGraph GitHub Wiki

IEDB uses implicit, heuristic templates to provide a pragmatic modeling approach for surveying the literature over a large number of different experimental studies involving immunoepitopes.

T-Cell and B-Cell responses

These seem to fall into the same overall structure from examination of the tcell and bcell tables

These types and subtypes fall into the following high-level categories:

  • Administration in vivo
    • Administration in vivo to cause disease
    • Administration in vivo to prevent or reduce disease
  • Exposure without evidence for disease
    • Documented exposure without evidence for disease
    • Environmental exposure to endemic/ubiquitous agent without evidence for disease
    • Exposure with existing immune reactivity without evidence for disease
  • No immunization
  • Occurrence of <X>
    • Occurrence of allergy
    • Occurrence of disease
      • Occurrence of autoimmune disease
      • Occurrence of cancer
      • Occurrence of infectious disease
  • Transplant/transfusion
  • Unknown

On the IEDB Study Papers page, we are working through the example provided by Richardson et al 1998, PMID9499101 to examine an example Administration in vivo protocol as a way of looking at this. The goal will be to extend this simple model to include other protocol types.