Review Comment:
The Nagoya protocol on Access and Benefit Sharing defines the legal processes constraining both the collection and reuse of data in Biodiversity studies. To find information regarding the existing studies and the administrative processes to collect and reuse the data requires to retrieve information from various heterogeneous databases and documents. It becomes particularly complex if you consider studies in regions of the world for which sovereignty is unclear as in the case of the United Kingdom and their external territories.
The authors are presenting in this paper a proof-of-principle ontology designed to improve the current situation by supporting the implementation of the FAIR principles. The NagO ontology is a cross domain ontology which aggregates ontologies for biodiversity, geography together with additional terms describing organizational and legal aspects of the Nagoya protocol and the different stakeholders involved. The ontology follows the OBO Foundry principles making it interoperable with the biodiversity relevant ontologies and some geographical ontologies. These various ontologies are aligned under the Basic Formal Ontology which provide a consistent semantic and logical framework.
The paper is well written and provides a clear overview of the context and the problems the ontology aims to address as well as a seemingly exhaustive list of the relevant data sources. The general design process of the ontology is clearly described with few exceptions that I list as minor comments below. This ontology is definitely relevant for the domain. However, I have several major concerns regarding the article, the ontology and the related data.
First, although the context is clear, the work is not supported by concrete use-cases from the domain which would illustrate the purpose and usage of the ontology. In section 5, usage scenarios seems to be more prospective scenarios than real tested ones. This definitely undermines the presentation of the work. I would suggest the authors to revise the paper to add concrete examples of use which would ground the work in practice and provide more clarity on practical challenges and added value for non-expert readers. These examples should provide both the related competency questions used for the design and the associated SPARQL queries to support the testing and exploration for new users. These queries could be shared on the github together with the ontology. Based on my understanding of the ontology and the context, I would suggest the authors to enhance the content of the ABS Clearing House (accessible through an API) to provide the missing link between the external territories and their sovereign country. This could be done by leveraging the Nagoya Look Up service mentioned in the article and developed by the authors.
My second major concern is related to the ontology itself. A lot of effort has been put to import biodiversity related ontologies such as ENVO or even the NCBITaxon while the main addition of the ontology focuses on the legal and organizational aspects of the information as shown by the example patterns used to illustrate the ontology. The paper does not actually show the connection between this organizational aspect of the ontology and the biodiversity, geographical aspects. This raises the question of the relevance to import these ontologies. Furthermore, as the author acknowledge in the paper, the ontology is actually really narrow while its scope is quite large. I would suggest the authors to extend it to at least other European countries such as France or the Netherlands which also have external territories and to consider at least one conflicted area. This would definitely enhance the usefulness of the ontology and also extend the ontology for governmental and sovereignty concepts.
My second major concerns about the ontology is related to the shortcuts used to build this first version of the ontology. First these shortcuts are not clearly explained in the paper. To understand the explanations, one needs to be an expert on the BFO and RO ontologies. I would suggest to clarify the text. In addition, one of these shortcuts implies that signatory roles are actual subclasses of “Homo Sapiens”. First this is semantically and logically wrong and second I am wondering about the need to describe persons based on their species. Why not using existing terms to describe persons? This issue actually would have an impact on the usage in applications.
My final set of major concerns is related to the data. The author are sharing the ontology on github. However, it is hard to understand which ontology file to use. The readme provides little documentation on the various files (e.g nago-base.owl, nago-full.owl, nago.owl). A first analysis of the ontology using Protégé revealed that none of these file includes both the additional NagO classes and their associated instances. After some digging I found the “complete” ontology in the src folder named nago-edit.owl. This is really cumbersome for interested users. It seems this particular github structure is tightly linked with the use of the ontology starter kit. I would recommend the authors to revise their README file to provide more information about its content. In addition, the ontology has little descriptive metadata (missing authors, contributor, description of the scope, …) and the link http://obofoundry.org/ontology/nago is broken.
Although this work is of high interest to support the enforcement of the Nagoya protocol and the implementation of the FAIR principles within this context, as presented here the ontology is clearly incomplete and would benefit from additional work.
Minor concerns
- Section 2 – Background
o Page 2 – column 2 line 31: “NagO aims to connect the vocabulary of different topics”. Should be “vocabularies” as many are existing and considered
o Page 3 – column 1 line 3-7: “The challenge is transforming (…) the observed data discrepancies”: this sentence is unclear, please rewrite
- Section 3 – Approach
o Page 3 – column 1 line 13 – “NagO was created (…)”: it follows the OBO principle and leverage the logical framework of the Basic Formal Ontology as Top level ontology. You should also mention that the ontology has 5 branches here.
o Page 3 – column 1 line 23-24 – “A general overview was developed and then extended with document terminology as well as Nagoya protocol processes and involved instance”: A general overview of what? What do you mean by document terminology?
o Page 3 – column 1 line 33 – “Very few of the necessary vocabulary (…)”: replace vocabulary by terms
o Page 3 – column 2 line 1-3 – “The consensus between all researched (…) as an overlap of definitions”: this sentence is not clear. Could you please rewrite?
o Page 3 – column 2 line 24-25 – “In order to put these terms into perspective they were looked at in a broader context”: Are referring to the people, document and process? This sentence is not clear to me. Please clarify.
o Page 3 – column 2 line 44-48 – “However, the ABS Clearing House itself provides openly accessible information on the role of genetic material (…) protocol text.”: What do you mean by “the role of genetic material”?
- Section 4 – Key features
o Page 4 – column 1 line 3 – “ There are five branches which NagO consolidates: (…)”: As the authors used already the term of branches for the describing the different part of the ontology, I would suggest finding a new term. This is confusing to the reader.
o Page 5 – column 1 line 30 – column 2 line 4 – “In summary, NagO combines (…) for business stakeholders”: this sentence does not make sense. Please rewrite.
o Figure 3 – shortcut relation is missing.
o Page 6 – column 2 line 11 – “(…) have been chosen very consciously”: replace by cautiously?
o Page 6 – column 2 line 12-15 – “For example, while (…) as these axioms”: this sentence is unclear. Please reformulate.
o Page 6 – column 2 line 16 – “Rather than creating BFO roles for terms”: what do you mean?
o Page 7 – table 1 – The table is not mentioned in the article. It can be removed.
- Section 5 – Usage scenarios
- Section 9 – Annex: the list of concept seems to be incomplete.
|