#767 term:transcription factor binding site identification

[jzheng]

http://purl.obolibrary.org/obo/OBI_0000291
transcription factor binding site identification

is defined as:
has_specified_output some (measurement datum and (is about some (gene expression and sequence-specific DNA binding)))

which cause inconsistency when term imported in BCGO ontology due to intersection of (gene expression and sequence-specific DNA binding) is null.

Shall change to (gene expression or sequence-specific DNA binding)?

[Alan Ruttenberg]

drop gene expression in the conjunct in specified output. I don't see how tf site identification is about gene expression other than indirectly.

Even if you wanted the datum to be about gene expression, you wouldn't want the "or". Instead you would write:

has_specified_output some (measurement datum and (is about some gene expression) and (is about some sequence-specific DNA binding))

i.e. there are two targets of is_about. "or" would mean one is satisfactory, and gene expression alone would not be sufficient.

[jzheng]

Thanks for correcting it. I was planning to edit my submitted tracker as you suggested before.

[Yu Lin]

Would the label for this term be too broad or ambiguous?

I have examples of computers predict the transcription factor binding site.
Or should I use label such as 'transcription factor binding site prediction' for my example, and use 'identification' for only experimental assays?

[Alan Ruttenberg]

It's on the edge. I think it could be clarified better by calling "transcription factor binding site identification" -> "transcription factor binding site assay"

[jzheng]

Discussed on OBI dev-call 20150615. We will change the logical axiom as what Alan proposed to fix the reasoning issue. However, the term need to be revisited due to the reason Alan pointed out on 2015-05-21 post.

I will also relabel the term to make it clear based on Alan's suggestion.

[jzheng]

changes were made in SVN revision 3992

The term will be revisited during OBI assay harmonization.