Why does ctd-to-owl add a gene expression node here?

[gaurav]

We were looking at http://ctdbase.org/detail.go?type=relationship&ixnId=3204472 on the call today

In CAM-KP, we see the following edge and its inverse as expected:

• CHEBI:15940 "nicotinic acid" biolink:affects NCBIGene:19 "ABCA1"

However, there is also a second edge that doesn't really make sense here:

• GO:0010467 "gene expression" biolink:has_participant NCBIGene:19 "ABCA1"

Is that useful to include? In TRAPI we're going to end up with a ton of genes that are recorded as participants of gene expressions, that I don't think are going to add much value.