Taxon constraint: adhesive extracellular matrix(GO:0062130) to Drosophila adhesive extracellular matrix

[DaiqingChen]

Please provide as much information as you can:

• GO term ID label:
  Replace ID label Adhesive extracellular matrix (GO:0062130) by "Drosophila adhesive extracellular matrix"

• Request to add a taxon constraint:
  Drosophila melanogaster

• • **only in taxon: **
    Drosophila melanogaster
• • **never in taxon: **

• Request to remove a taxon constraint:
  The old term is misleading as in common usage the term "adhesive extracellular matrix" refers to the group of gene products that exert adhesive functions within the extracellular matrix (GO:0031012), while this term is actually defining a subset of drosophila gene products; the reference says "These glands synthesize and secrete massive amounts of a glue which attaches the pupae to the substrate during metamorphosis.", this is taxon-specific.
  We thus recommend changing the name to "drosophila adhesive extracellular matrix" or the synonym "puparial glue" and also change the parent term.

• Change the parent to Specialized extracellular matrix (see Add term Specialized extracellular matrix #29265)

• Supporting evidence if available (e.g PMID):
  Keep the same reference and definition: PMID:825230

[Matrisome]

@DaiqingChen : Add link to slide deck and insert screenshot of revised slide 2

[cmungall]

Hi @DaiqingChen

You can find some of the original discussion that led to the creation of this term here:

• Fly glue term #17502

Interestingly, the original request from @hattrill was for a Drosophila-specific term, but after discussion we decided to introduce a very broad and general term.

Note also that it used to be common practice to attach the PMID that was used for the first annotation to the term definition, even if the PMID is much more specific. This is really confusing. We have since stopped this practice.

I agree with your broader analysis. The current term as it exists is far too broad, as it covers a huge range of secreted products that are different structurally and evolutionary. I support "repurposing" this term to be more specific as it was originally intended.

I think this is a good time to do some general cleanup. Let's look at some similar terms:

• GO BP term "puparial adhesion" https://amigo.geneontology.org/amigo/term/GO:0007594
• FlyBase gene group term "glue protein" https://flybase.org/reports/FBgg0001189.htm

The gene sets for the 3 terms are identical (modulo inclusion of Sgs9, which was purported to exist in 1976 but has not been genomically characterized)

It seems odd to me that we have 3 terms to describe the same set of genes, yet in GO we report "function unknown" for most of them. The biophysical function of these genes are understood in exquisite detail! The function for some of the genes is reported as "structural constituent of ECM" but this is inconsistently applied and IMO is pretty indirect.

I think the most conservative course of action here is to repurpose this term as you suggest in #29265 (comment) under "specialized ECM", relabel and redefine to be less generic, add a taxon constraint. I think the correct TC is NCBITaxon:480117 ! Cyclorrhapha (see https://link.springer.com/chapter/10.1007/978-3-319-40740-1_15) but putting "Dipteran" in the name is good enough for clarity.

I would make an equivalence xref to FlyBase:FBgg0001189 such that gene sets are guaranteed to be the same (though this seems fairly stable)

And not to overload this ticket, but I'm questioning the utility of:

id: GO:0007594
name: puparial adhesion
namespace: biological_process
def: "The adhesion of the puparia of Diptera to their substrate; normally effected by a 'glue' secreted by the larval salivary gland and expectorated at the time of pupariation." [GOC:ma]
synonym: "puparial glue" RELATED []
is_a: GO:0022609 ! multicellular organism adhesion to substrate
relationship: part_of GO:0007591 ! molting cycle, chitin-based cuticle

Is this truly a biological process, or more of a biophysical process? I think if we do want to represent the GO-BP aspect of glue genes we should have a term for the secretory pathway for generating them, which is well understood with defined starts and ends (though the glue genes themselves would not be annotated here). It seems odd to try and force the glue genes into a biological process.

I also wonder if we want to do a better job of gene products whose function is biophysical. We have the "structural constituent" branch but this seems to be applied consistently.

(oh and the reference for the FBgg0001189 is Naba et al, we have come full circle :-)

[Matrisome]

@cmungall Thanks for the feedback!
Your proposal seem reasonable. If we need additional insight, I can reach out to our ECM Drosophila expert.

[ValWood]

I'm guessing this doesn't require a taxon constrain. Can we close?