SBOL 2.2.1 Request for Comments

[Chris J. Myers]

Hi,

During HARMONY, we identified a number of small issues that needed correcting in the current version of the specification.  We have since made these corrections, and we are proposing to release the SBOL 2.2.1 specification at the end of July.  This is attached for your further comments before release.  If you discover any problems, we encourage you to log them to the SBOL specification issue tracker:

https://github.com/SynBioDex/SBOL-specification/issues

All significant changes for this version are marked with a 2.2.1 label in the margin near the change (minor typo fixes may not be noted).  These changes include:

1) A new TopLevel UML diagram that includes all new TopLevel classes.

2) A clarification on ontology terms for roles for proteins and small molecules.

3) A best practice for annotation of authorship.

4) A new example showing how Prov-O should be used for designs derived from CombinatorialDerivations.

5) A lot of corrections to the validation rules.

A full list of the 41 issues addressed can be found here:

https://github.com/SynBioDex/SBOL-specification/issues?q=is%3Aissue+milestone%3A%22Version+2.2.1%22+is%3Aclosed

The updated validation rules are also now supported by the latest version of libSBOLj (2.3.2-SNAPSHOT), as well as the online validator.  The plan is to release this version of the specification and libSBOLj at the end of July.

Thanks,

Chris

[Hiroyuki Kuwahara]

Hi Chris,

This is probably not related to spec 2.2.1, but I have a question regarding 2).  What's the reason behind the decision to have Effector as a role of SmallMolecule ComponentDefinition? 
To me, this kind of roles is specific to the context of biochemical reactions in a system you want to describe, and I think Effector is more appropriate as a Participation Role, not as a ComponentDefinition role.

ChEBI defines Effector as "A small molecule which increases (activator) or decreases (inhibitor) the activity of an (allosteric) enzyme by binding to the enzyme at the regulatory site (which is different from the substrate-binding catalytic site)."
And in ChEBI, Acetyl-CoA is defined to have role Effector.   So, as a best practice, Acetyl-CoA CompDef should have this Effector role, right?  But, say, if I want to describe biosynthesis of a natural product with Acetyl-CoA as a precursor, then I don't think it makes sense for Acetyl-CoA to have this role in this specific context.    I think this point is different from the other roles defined in Table 4.  Promoter means a promoter, and this stretch of DNA is intended to be used as a promoter in the system you want to describe.  

thanks,

hiro

--
You received this message because you are subscribed to the Google Groups "SBOL Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sbol-dev+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "SBOL Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sbol-dev+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Chris Myers]

Hi Hiro,

I think you are right that these terms can be context dependent.  Indeed, we are just starting to get our heads around the use of rule for non-DNA components.  I believe this particular entry has been in the table for awhile though.  I think the way to think about role is that it gives a hint of the likely function of a component.  Even in the case of a DNA component, the same sequence could have a very different role in a different context.  For example, if you have a stop codon component, the same bases could appear somewhere else outside a CDS and have a very different function.  I suspect there are better examples too.  

The primary use of a role is in filtering when you are looking for a component that can serve a particular purpose.  For example, give me a list of all promoters.  So, the question becomes, would you like to see Acetyl-CoA in a list of small molecules that can serve as Effectors.  If so, then I would suggest that it should have this role.  The role is a list, so if Acetyl-CoA can also have other roles, then it should be okay to include those other roles as well, so it would say also appear in a list of potential precursors.

Does this make sense?  Anyone else has comments on this?

Cheers,

Chris

To unsubscribe from this group and stop receiving emails from it, send an email to sbol-dev+u...@googlegroups.com.

[Hiroyuki Kuwahara]

Hi Chris,

Thanks.  I understand your point, and I noticed that the spec also had a statement which is in line with what you said: "The roles property is an OPTIONAL set of URI s that clarifies the potential function of the entity represented by a ComponentDefinition in a biochemical or physical context."  So, yes, it makes sense.

But in this case, I think a large number of small molecules should have this effector role since they can potentially be used as allosteric regulators of some engineered ribozymes.  And I think this specific role becomes meaningless.

Anyway, thanks.

hiro