Disease Specific Rules

From curation_manual
Jump to: navigation, search

Infectious Diseases

General Considerations

When curating epitopes derived from infectious agents, the immunogen in the assay is always relative to the antigen in the assay. Typically, this will be the source organism of the epitope or a taxonomic relation to the source organism. The main question to be asked for all infectious disease assays is were the subjects exposed to or infected by the source organism?

Exposure without Evidence for Disease-Environmental Exposure to Endemic / Ubiquitious Agent

This selection is to be used when there is inferred exposure to a pathogenic agent, by merit of its endemicity or ubiquity, but no information on the disease outcome, seropositivity, or actual events of exposure. This includes pathogens that have a high incidence in the population (Influenza, CMV, EBV, Candida, etc). Examples are EBV exposure in healthy subjects and P.falciparum exposure in an endemic geographic area.

   * The disease name and stage should be left blank. 

For example: If the epitope is derived from influenza A PR8 and the host of the assay is a human, was that human exposed to or infected by influenza A? The IEDB considers all humans to have been exposed to influenza, so unless the authors state otherwise. The in vivo process will be curated as "Exposure to ubiquitous agent" with immunogen "Taxonomic parent" = Influenza A virus.


In infectious disease, controls subjects that have not been exposed to or infected by the infectious agent from which the epitope is derived, are not curated unless they are positive. If some infectious disease epitopes are positive and some are negative in the healthy population, and the number of epitopes is large, the curator may opt to curate all epitopes, if this saves time, but only the positive epitopes require this context be curated.


Patients having tuberculosis are often described using a variety of disease stages in manuscripts. The IEDB uses the following criteria to curate hosts having tuberculosis.

Occurrence of infectious disease

  • Patients described as having active or symptomatic mycobacterium disease (recent, remote transmissions) will be curated as disease stage of chronic unless the text specifically states as acute.
  • Patients described as having had prior TB, as inactive, latent, treated and resolved, or asymptomatic disease should be curated as disease stage of post.

Administration in vivo

  • Subjects that are described as vaccinated (BCG, etc) will not have a disease state or stage filled in.

Exposure without evidence of disease

  • Subjects that are described as having exposure, but without conversion/seropositivity

Exposure without evidence of disease with immune recativity

  • Subjects that are described as having exposure and are PPD+


SARS-CoV2 has several specific rules that differ from other infectious disease curations.

Isolates and strains

Make IEDB term requests for isolates and species that are children of Severe acute respiratory syndrome coronavirus 2 (2019-nCoV) (ID:2697049, SARS-CoV2). For SARS-CoV2, we are requesting every strain/isolate/variant studied as the epitope source, the immunogen, or the antigen. Note this is an exception that is being made just for SARS-CoV2. Term requests MUST use correct variant parent, such as SARS-CoV2 Alpha, NOT SARS-CoV2.

  • If the paper gives a specific strain anywhere in the manuscript, that should be requested and used in curation
  • If a recombinant S protein (or RBD, NTD, S1, S2 fragment) is made with the mutations defining a lineage (B.1.1..7, B.1.351, etc), use that variant rather than a specific strain.
  • If the paper only uses a variant name (B.1.1..7, Alpha, B.1.351, Beta, etc) and does not provide a specific strain, use the variant rather than looking up what strain they mean.

Assay Type

ACE2 inhibition should be curated as neutralization assay


For coronavirus papers, curate exposed, infected, and healthy donors whether the outcome is positive or negative. Healthy donors are exposed to ubiquitous coronavirus.

Immunogen and Antigen

  • Pseudovirus should be curated as protein, not organism
  • If the antigen tested is RBD or the S1 domain of spike, this is Fragment of Source Ag (not whole protein). If the authors tell you exactly what fragment they used, enter that. If not use RBD (R319-F541) or the S1 (V16-R685).
  • If they test both wild type organism or protein and variant organism or protein, capture as two separate contexts (wt & variant)
  • If authors provide specific strain information for the variant, use that (tax sibling, tax child, etc)
  • If wild type and variant have same tax id, for ex, both are SARS-CoV-2, then use other structure from source org
  • If variant is a fragment or a protein, try to use an accession with the variant in its name, for ex, “SARS-CoV-2 D614G 3 RBD down Spike Protein”

https://www.ncbi.nlm.nih.gov/protein/?term=D614G Or use blast search


All humans are considered exposed to common allergens, for example to pollens, cat dander, etc. For this reason, in addition to the allergic subjects, healthy controls are always curated for allergy contexts. Healthy donors who are exposed will be curated as "Environmental exposure to endemic/ubiquitous agent without evidence for disease" (or documented if more appropriate). Healthy donors who are not exposed to the allergen will be curated as "No Immunization" because their reaction or lack thereof is relevant.


General Considerations

When curating epitopes from autoimmune related antigens, all subjects express their self antigens, therefore, we also curate healthy controls. However, we do not curate an immunogen in the in vivo processes. Instead in vivo process 1 captures either "Occurrence of Autoimmune Disease" for the study subjects or "No immunization" for the healthy controls. When no additional immunization occurs, all in vitro procedures are considered "primary induction in vitro".

In references that describe several different study subjects with different autoimmune diseases, capture the group relevant to the epitope separately. For example, if the epitope is from insulin, the diabetes group is the relevant group. Other groups may be curated separately or bulked together depending on the assay outcomes and the relevance of the disease groups.

If there is uncertainty whether a disease is an autoimmune disease or not, open the Disease Tree and find the parent of the disease in the tree. If the parent term is or is a child of 'autoimmune disease', then the appropriate in vivo process is "Occurrence of Autoimmune Disease". If the parent term is not a child of 'autoimmune disease', then the in vivo process must not be "Occurrence of Autoimmune Disease".


Diabetic Subjects

In vivo Process Type: Occurrence of Autoimmune Disease

Diseases state: Most are IDDM [E10] Latent autoimmune diabetes in adults (LADA), pre-diabetic, etc can use ICD 10 R73.0 (Abnormal glucose tolerance test)

Disease Stage: Acute will be edited to read Acute/recent onset Use author comments to determine acute vs chronic

Immunization comments: Very important to note everything the authors state regarding the subjects (C peptide +/-, antibodies, time since diagnosis, etc).

Immunogen: NONE


Curate all controls, positive or negative In vivo Process Type: No Immunization

Diseases state: Healthy if they are, Blank if not & bulking several diseases. Capture non-diabetic diseases only when they are relevant.

Immunization comments: Bulk all controls with the same outcome, even if they have different diseases (RA & Adrenal insufficiency for ex).

Immunogen: NONE

  • Special Note: Because both diabetic and control subjects have no immunogen, all in vitro processes will be “Primary induction in vitro” even though the subject may have been exposed in vivo to the self antigen. We are considering adding a new in vitro process type &/or changing the way external represents these cases.

Animal Models

If the animal has diabetes, curate the disease state as IDDM (use ICD-10). If the animal has a disease that you cannot find, bring it up & request that it is added. If a certain mouse strain is needed, request that it is added per the usual process. If the animal model is prediabetic, curate the animal as "Occurence of Autoimmune Disease" using "abnormal glucose tolerance test".

Alzheimer’s Disease

This disease is the most common form of dementia. It is characterized by neurofibrillary tangles inside nerve cell bodies and amyloid plaques in the brain. The plaques are composed of peptides from the amyloid precursor protein, a normal cellular component.

We are not curating papers with only immunohistochemistry with mAbs. But, if a reference has that, along with other curatable data, then curate all of the paper.

If they show that an antibody stains human (or APP mouse) brain plaques, the antigen = amyloid precursor (source antigen)

Celiac Disease

Celiac disease is considered an autoimmune disease by the Disease Ontology, therefore use in vivo process type of Occurrence of autoimmune disease and enter the disease state of celiac disease. For the immunogen, select Derivative or Organism = gluten.

Human Subjects and Mice

In vivo Process Type: Occurrence of Disease

Diseases state: Most are: Alzheimer's disease [G30] For mice, the model for the disease is APP or APPsw transgenic mice. The disease state for them is: Alzheimer's disease [G30]

Disease Stage: Usually chronic for humans

Healthy Controls

In vivo Process Type: No Immunization

Diseases state: Healthy Immunization comments: Bulk all controls with the same outcome, even if they have different diseases. Comment on the diseases and demographics of these groups when you bulk.

Immunogen: NONE

  • Special Note: Because both Alzheimer’s Disease and control subjects have no immunogen, all in vitro processes (when present) will be “Primary induction in vitro” even though the subject may have been exposed in vivo to the self antigen.

Immunogen: NONE

Amyloid beta 1-42 Epitope Sequence Exception

A common epitope is Amyloid beta 1-42 (A beta 1-42). The standard sequence is:


If no sequence is given in the paper and you are sure that they are using this standard sequence (for instance, they purchased the peptide), you should use the above sequence and the following: Data Location = “Standard reference sequence” Epitope evidence code “representative” If Immunogen or antigen = epitope, Imm/Ag evidence code = “representative” b/c the sequence of the peptide is representative

You may only do this for this epitope. For any other epitopes you encounter, you must verify sequence info as usual.


see http://curationwiki.iedb.org/wiki/index.php?title=CEDAR/Cancer_Specific_Rules&action=edit&redlink=1


Transplant Categories

The following figure demonstrates some of the categories of papers that we curate:


  • Alloreactivity is the reaction between the immune cells of one individual and the components of another individual
  • Minor Antigens are recognized after allogeneic hematopoietic stem cell transplantation
  • Allo-Peptides are presented by MHC molecules not present in the host
  • H-Y Protein Female T cells are capable of recognizing peptides derived from H-Y proteins following transplantation into male recipients
  • Xenoreactivity The reaction between the immune cells of one species and the components of another species.
  • Graft vs Host Disease (GVHD) The transplanted components of the immune system attack the recipient of the transplant.

Basic Rules

The most important points to understand when curating transplant are where are the effectors from and how did the host become exposed to the epitope containing object. When curating any epitopes, always consider what the host was exposed to in order to mount an immune response to the assay antigen.

As always, if the epitope is derived from influenza and assay antigen is epitope, source antigen or source organism, the immunogen is influenza.

If the epitope is derived from a protein present on donor tissue, and the host’s immune system mounts an immune response to the transplant/transfusion, the immunogen is the transplanted tissue or blood cells/products.

When curating a transplantation or transfusion as an in vivo process, the immunogen is the tissue, cell, or protein that was transferred. For example, cardiac transplant immunogen = Derivative of Source Organism/Anatomical Entity, Tissue, Tissue Type = Heart, SOrg = homo sapiens.

Stem cell transplant immunogen = Derivative of Source Organism, Cell, Cell Type = stem cell, tissue type = Bone marrow, SOrg = homo sapiens.

You always must have Immunization Comment describing the transplant process and any relevant information. For example, “A male recipient received a stem cell transplant from a female donor.”

For example, a recipient makes antibodies to a protein present on a donor’s tissue. Lung transplant leads to reactivity to donor’s HLA-DR epitope present within the lung tissue.

Epitope is derived from HLA-DR

Curate the transplant/transfusion:

In vivo 1 = Transplantation

IV1 Immunogen = human lung

Immunization comment = “The HLA-DR negative recipient received a lung transplant from an HLA-DR positive individual.”

Disease State

Only capture the disease that requires the transplant/transfusion as in vivo process 1 if it is relevant to the immune recognition of the antigen.

Thus, Factor VIII deficiency is captured when the epitope’s source antigen is Factor VIII and drug overdose leading to the need for a liver transplant is not captured. In the case of Factor VIII deficiency, the host does not naturally express Factor VIII, making this disease relevant. In the case of drug overdose, this disease has nothing to do with why the host's immune system mounted an immune response.

Adoptive Transfer vs Transplantation for Stem Cell Transplants

  • When the effector cells themselves originate from one host and are transplanted into another host via a stem cell transplant, this is curated as adoptive transfer.

  • When the stem cell transplant process does not supply the effectors, but instead is the source of the epitope containing material (source antigen or cell expressing the epitope), the stem cell transplant process is curated as in vivo process type = Transplantation/Transfusion.

Hemophilia Rules

In cases where a person makes an antibody to a human protein/epitope as a result of being administered a transfusion product (blood, platelets, fake blood, Factor VIII, etc) containing that same protein/epitope.

For ex, a subject does not make Factor VIII (Factor VIII deficiency) then he gets several Factor VIII treatments and subsequently develops Abs to Factor VIII.

IV1 process = occurrence of disease

IV1 disease = Factor VIII deficiency

IV2 process = administration in vivo

IV2 Immunogen = Factor VIII (Source antigen of epitope)

Immiz comment must explain that the subject(s) received transfusions.

alpha GAL Epitope

Alpha-Gal-1-3-beta-Gal is a disaccharide found (usually ligated by 1-4 linkage to GlcNAc) on glycolipids of the cell membrane as well as a part of N-glycans of proteins. It is expressed widely in the animal kingdom and in some microorganisms. It is not expressed in Humans and Old-World monkeys. The alpha-gal epitope and galactosyl should not be confused with alpha-gal ceramide. The latter has only one carbohydrate unit (galactose) linked to ceramide and interacts with receptors on NKT cells.

The alpha-Gal epitope is not found on humans and some monkeys, but is found in pigs and rabbits. It is the main xenoantigen hampering the success of clinical xenotransplantation. It can be synthesized and tested with no reference to an organism or can be used only in the context of an organism.

Epitope ChEBI ID

ChEBI has multiple possible entries. When no specific detail is given for the structure, use "α-Gal epitope (with structure) CHEBI:59172" = A trisaccharide antigen comprising one N-glucosamine and two galactose residues, linked as shown; found in all animals apart from Old World primates and humans. The main xenoantigen hampering the success of clinical xenotransplantation.

When the authors do provide a structure, ChEBI will be requested to view the manuscript and assign the correct ChEBI ID.

Source Antigen and Organism

Unless the authors specifically extract the epitope from a specific source antigen, curate with no source antigen. If the assays utilize the source antigen of the epitope as the immunogen or the antigen, then you should curate the epitope with a source antigen.

Always curate the source organism as specified by the authors. In some cases, this will result in more than one curated epitope.

When curating the epitope without a source antigen, it will be an accession non-sequence molecule. When curating the epitope with a source antigen, it will be a fragment of a natural non-sequence molecule.

Example: rabbit cells are used to make a mAb to aGal that binds to pig cells.

Curate 1 epitope with source organism = rabbit and 1 epitope with source organism = pig. Curate the assay under both.

Example: Mouse is immunized with pig kidney membranes to generate polyclonal antibodies to aGal.

Curate 1 epitope with source organism = pig.

Immunogen and Antigen

Curate what is transferred, injected, transplanted or tested that contains the epitope. Typically, this will be the tissue or cell that expresses the epitope. In some cases a purified epitope may be utilized or curating the epitope plus carrier may be needed to clarify how the experiment was performed. For cells and tissues naturally expressing the epitope, curate the immunogen/antigen as the cell or tissue. For engineered or treated cells, curate the immunogen/antigen as epitope.

Do not curate natural exposure to gut flora as the immunization process for animals exposed to aGal. Only curate exposure as a result of transplant or administration of tissue/cells containing aGal.

Transplant Rejection Assays

When treatment with epitope (or epitope specific cells/antibodies) leads to graft rejection or graft acceptance, this is curated as Reduction of disease after treatment or Exacerbation of disease. The assay antigen is the tissue or cells that were transplanted and rejected/accepted.


Some manuscripts that describe exposure to a foreign antigen expressed by a fetus, but not by its mother have been included in the Transplant Category.

In these cases, the pregnancy results in the exposure to the immunogen.

The in vivo process type should be curated as "Exposure without evidence for disease" and the immunogen should be the Embryo. Be sure to use immunization comments to explain the scenario.