OBO Foundry Food Ontology Interconnectivity

Tracking #: 3305-4519

Damion Dooley
Fleur Mougin1
Liliana Andrés-Hernández1
Carl Lachat
Francesco Vitali
Georgeta Bordea
Magalie Weber
Hande Kucuk McGinty
Chen Yang
Pol Castellano-Escuder
Leigh Carmody
Duccio Cavalieri
Lauren Chan
Matthew Lange

Responsible editor: 
Guest Editors Global Food System 2021

Submission type: 
Survey Article
Since its creation in 2016, the FoodOn food ontology has become an interconnected partner in various academic and government projects that span agricultural and public health domains. This paper examines recent data interoperability capabilities arising from food-related ontologies belonging to, or compatible with, the encyclopedic Open Biological and Biomedical Ontology Foundry (OBO) ontology platform, and how research organizations and industry might utilize them for their own projects or for data exchange. Projects are seeking standardized vocabulary across many food supply activities ranging from agricultural production, harvesting, preparation, food processing, marketing, distribution and consumption, as well as more indirect health, economic, food security and sustainability analysis and reporting tools. To satisfy this demand for controlled vocabulary requires establishing domain specific ontologies whose curators coordinate closely to produce recommended patterns for food system vocabulary.
Full PDF Version: 

Minor Revision

Solicited Reviews:
Click to Expand/Collapse
Review #1
Anonymous submitted on 15/Mar/2023
Major Revision
Review Comment:

This review addresses Review #2 from the previous revision.

First, off thank you for your response to my comments. However, I find them to be lacking appropriate detail to make an improved judgement. Perhaps this is reasonable since my own comments are quite top-level. On the other hand, I also see that the response to my review is unfinished.

The manuscript is still, unfortunately, quite disjointed. It's also still missing a strong identity. The introduction does not have an explicit list of what contributions this manuscript makes. The discussion section does not have an exhaustive address of next steps or meaningful way to accomplish filling of identified gaps. (Also, I don't think that, in general, design patterns are known as nanopublications).

What is the purpose of Section 2? I would guess this should be a /governance/ section. And in that, it does not seem to be exhaustive. It is missing the canonical citation to protege (and it should be capitalized as well). I see that MIREOT is motivated, but not really explained. It's mentioned to be a minimum requirement, but you follow up that not everyone uses it.

The paper still suffers from significant formatting and presentation issues.
* The article does not conform to proper iosart style
* The Abstract is the wrong font and is not justified
* at least one affiliation is out of date.
* The first table should probably be an appendix. It also does not conform to the proper table style for iosart
* heading font and size are incorrect
* spaces should appear before a citation (this is inconsistent anyway, across the manuscript)
* Figure 1 does not have a thorough explanation, still. What does bidirectional import really mean? What are non-dotted arrows?
* third, 3rd, 3^rd
* inconsistent use of UK vs American English styling "e.g.," vs "e.g."
* There are, in bright red and caps, missing references.
* Captions of figures are inconsistently centered, have the wrong font sizes.
* Pages 7, 10, has a ton of open space for some reason
* Inconsistent usage of your defined acronyms.
* Inconsistent usage of styling for uris, footnotes, namespaces, references to data. There is at least one url styled as a hyperlink.
* References are not formatted properly

Overall, I find that the paper suffers from exactly the same (content) problems as the first review, and I maintain that significant improvements are necessary.

Review #2
By Chris Mungall submitted on 15/Mar/2023
Review Comment:

Thank you for the revised version, this is much improved in terms of
coherency and consistency of sections. This is now a fantastic summary
of the challenges involved in interoperating across a range of
ontologies in overlapping domains.

I have only minor suggestions:

1. MiXS should be MIxS

2. Figure 1:

Figure 1 still has issues. NCBITaxon and Uberon are both attributed to
my institute but in fact Uberon is a multi-institute collaboration,
and we certainly don't make the NCBI taxonomy!

I don't think PO came from the college of public health.

But more generally it is odd to attribute ontologies to institutions,
most ontologies are inherently team science multi-site collaborations
with a vast range of global contributions.

3. There are various pieces of red text that say "REFERENCE" which
presumably need to be filled in.

Review #3
Anonymous submitted on 04/Apr/2023
Minor Revision
Review Comment:

(1) This paper is suitable as an introductory text to get one started on the covered topic, but there really should be a subsection in section 3 specifically for a description of FOODON. A new person to this topic will need that additional information.
(2) This paper is comprehensive and balanced in its presentation and coverage, except for the need for a subsection on FOODON.
(3) This paper is very well written and clear. Figures are clear and useful.
(4) This paper is important to the broader community, as no such comprehensive paper on this subject exists. I learned a great deal by reading it and expect the information in this paper to be very useful to a wide audience. Each of the specific ontologies mentioned in this paper are accessible via different publicly accessible routes.
Minor comments:
Disease Ontology (DO) is not in the acronym table.
The use of the term "curation" for ontology development is not standard. I've been affiliated with the International Society for Biocuration since its inception and have never heard anyone describe ontology building or development as curation. I also have not experienced anyone in the OBO foundry calls, nor at the ICBO conference using this term in that way. That is not to say that it is definitely incorrect usage, just that it is not common usage. The OBO Foundry is not known as a "curation consortium". If you visit their "about" webpage and the https://en.wikipedia.org/wiki/OBO_Foundry page, the word "curation" cannot be found.
NCBI stands for "The National Center for Biotechnology Information" not "National Center for Biocomputing Technology" (in text and acronym table).
Introduction, 2nd paragraph, insert the word "of" between "overview" and "food".
Introduction, 2nd paragraph, insert the word "the" between "what" and "OBO foundry."
There is inconsistent use of a space before brackets. Sometimes there is a space and sometimes there is not.
The ROBOT tool should be written as "ROBOT" not "robot."
There are 2 places where the red text "[REFERENCE[" was left in, rather than being replaced with the correct reference number.
In the Discussion, in the 5th paragraph, the word "Vitamins" is erroneously capitalized.

Review #4
By Adriano Di Pasquale submitted on 05/May/2023
Minor Revision
Review Comment:

This paper shows an extensive number of domain food ontologies and their interconnectivity within the framework of the OBO Foundry.
It investigates (i) technical aspects on how it is possible to integrate them, (ii) organizational aspects on how to manage synchronizations and (iii) stakeholders aspects, such as working group organizations.

In the last version, the manuscript has been improved a lot and the content harmonized.

The main concerns are:
- the difficulty to catch usage examples in so many disparate cross-sectors ontologies,
- which is exactly the final "set of interconnected" ontologies. It seems it is a kind of "greedy approach" to ontologies integrations which might be confusing for different stakeholders. For instance, for genomic epidemiology stakeholders, some ontologies are interesting (e.g., foodon, hso, ) while others no. Probably the same applies to other stakeholders.
- Clear examples for different stakeholders. Indeed the Selenium examples, in my opinion, do not reveal the power of integrations for genomic epidemiology stakeholders.

I think that some other examples might help (e.g., pathogens found in food during inspections). Probably could be inserted as supplementary materials without the need to change the main text.

In the supplementary materials, it would be helpful to have source code (e.g., files that can be opened with protege software) in order to evaluate the integrations.

With the suggested supplementary materials, the paper would be improved and should be accepted.