Human Inborn Errors of Immunity: 2022 Update on the Classification from the International Union of Immunological Societies Expert Committee

[pnrobinson]

https://link.springer.com/content/pdf/10.1007/s10875-022-01289-3.pdf

Can we use this as the backbone for the MONDO classification of these diseases?

[matentzn]

I will take that on with a medical expert I work with. Here is a snippet of the data they provide:

Disease	Genetic defect	Inheritance	OMIM	T cell count	B cell count	Immunoglobulin levels	Neutrophil count	Other affected cells	Associated features	Major category	Subcategory	ICD9	ICD10	HPO (table)	HPO (subtable)
CD3d deficiency	CD3D	AR	186790	Very low	Normal	Low	Normal	No g/d T cells		Table 1 Immunodeficiencies affecting cellular and humoral immunity	Subtable 1 T-B+ SCID	279.2	D81.2	HP:0002715	HP:0005403
CD3e deficiency	CD3E	AR	186830	Very low	Normal	Low	Normal	No g/d T cells		Table 1 Immunodeficiencies affecting cellular and humoral immunity	Subtable 1 T-B+ SCID	279.2	D81.2	HP:0002715	HP:0005403
CD3z deficiency	CD3Z	AR	186780	Very low	Normal	Low	Normal	No g/d T cells		Table 1 Immunodeficiencies affecting cellular and humoral immunity	Subtable 1 T-B+ SCID	279.2	D81.2	HP:0002715	HP:0005403
Coronin-1A deficiency	CORO1A	AR	605000	Very low	Normal	Low	Normal		Detectable thymus, EBV	Table 1 Immunodeficiencies affecting cellular and humoral immunity	Subtable 1 T-B+ SCID	279.2	D81.2	HP:0002715	HP:0005403

The question I have:

1. Should we create grouping classes for all "tables" and "subtables" (corresponding to HPO phenotypes!) in that paper? If not, what should be used as the defining feature (given that table) for the classification? The genes?
2. I will start by my collaborator reading the paper carefully, linking it to what is currently in Mondo via the OMIM mappings, making sure that the Mappings are correct according to the paper.
3. The next step is to decide how to classify (see question above) where we need your advice (we can start by assuming that we classify according to the tables (phenotypes) and according to the genes, and then decide later which way to go.

Please advice.

[sabrinatoro]

The classification in the paper is gene-based (which is how the ClinGen Expert panel(s) view disease classification. We can contact them if we have additional question).

Here is how I understand it:

• Each table represents a high-level grouping term. Each sub-table (title in blue) represents a parent term, and each row in the table represents a disease.
  For example, in Table 1:
• There are 3 high-level parent
  • T-B+ Severe Combined Immune Deficiency (SCID)
  • T-B- SCID
  • Combined Immunodeficiency (CID)
• This classification is based on affected genes
  • Each row represents a disease and should be represented by a Mondo term
  • These Mondo terms should correspond to the OMIM record reported in the table
  • The gene-specific name (“disease” column, ~ deficiency) should be added as a synonym for the Mondo terms if these names do not exist yet
  • When more than one OMIM correspond to a disease (e.g. RAG deficiency), a new parent term is needed to group these OMIM terms and be called by the disease name in column A (e.g. “RAG deficiency”)

I need to read the paper carefully to answer the following:
I am not sure that every table and sub-table should represent a specific branch in the ontology. For example, there are tables called “Other defects” or “defects of motility” or others referring to HP phenotypes. I think it might make sense in some cases, but not all.

We (Mondo curators/experts) should review these and decide how the high-level classification should be. I will start on that.

At minimum, ALL diseases (ie all OMIM) mentioned in all the tables should be in the Inborn errors of immunity branch.