susceptable_to relationship

[mbrush]

Initially reported on RO tracker: https://code.google.com/p/obo-relations/issues/detail?id=31 Original Reporter: Nicole Washington

Such a relation is needed in GENO as well, so we should work to get this into the RO and use in GENO

---

Initial post: we need a relationship that aids in describing the susceptibility of a disease due to variation(s) of genomic regions.

for example: http://omim.org/entry/607339

this entry describes variants in the genomic region of human chr16pter-p13 as contributing to the susceptibility of coronary heart disease.

this kind of relationship is necessary for relationships created through statistical associations.

---

Comment from: peternro

There are two different issues here.

1. Susceptibility:

contributes_to_susceptibility_for

e.g., if there is a GWAS hit rs1234 for disease X, then

rs1234 contributes_to_susceptibility_for 'disease X'

2. single (small) locus vs. genomic region rs1234 is something like chr7:12345654G>A that is, well defined position. This may be causative or it may be in linkage disequilibrium with an unknown causative variant.

On the other hand, a genomic region such as in the above OMIM entry is 800,000 bp long, was identified by linkage analysis ("statistical associations"), and the actual variant is simply somewhere in this region. Thus, future research is needed to find the actual variants that causative. The actual nature of linkage is different for Mendelian disease (long distances, family specific) and common disease (shorter distances, population specific). OMIM does not really model this stuff well. If we want to do this for Monarch, I would suggest that we do about 10 examples together in detail, these are things that confuse most MDs and biologists and you need to keep staring at them for a long time.

By the way, susceptibility and the locus-region issue are completely orthogonal to one another!

---

Comment from: nicole.l.washington

yep, i only intended this to be an issue concerning the susceptibility relationship, which i perceive to be between any kind of genomic feature (which could be a specific base pair, or a large arm of a chromosome).

the difference between direct causation vs one (or more) unknown variants in a region with this relationship could either be captured within the relationship, or it could be part of the evidence code.