|
| Moderated by Stephen Muggleton. |
Steffen Schulze-KremerOntologies for Molecular Biology |
The
article
mentioned above has been submitted to the Electronic
Transactions on Artificial Intelligence, and the present page
contains the review discussion. Click here for
more
explanations and for the webpage of theauthor, Steffen Schulze-Kremer.
Overview of interactions
Q1. Anonymous Referee 1 (15.10):
Recommendation: The paper can eventually be accepted for ETAI. Another refereeing round is needed. The paper introduces ontologies in general, but the overall description is too superficial for biologists or bio-informatics researchers to appreciate the usefulness of ontologies for molecular biology. The paper needs major revision according to the line suggested below: 1. The most part of Section 3.1 on muKosmos should be shorten while leaving the last paragraph that muKosmos does not include a lot of concepts from molecular biology. 2. In Section 3.2, the author ought to provide some examples from molecular biology, which will be much more appealing. 3. It is hard to appreciate that the ontology in Figure 4 is tailored to molecular biology. It would be very helpful to the ontology to re-formulate some specific terms used in, for instance dbEST or SwissProt. 4. In Figure 5, although a couple of molecular biology terms, such as GDB Genome and Linkage map, were mentioned, the table also ought to include more biological concepts. For instance, it would be interesting to incorporate STS markers, micro-satellite markers, and single nucleotide polymorphism into your ontology. Q2. Anonymous Referee 2 (15.10):
"Molecular biology has a communication problem" -- This problem is very real. Physics has Mathematics for its scientific language and Chemistry, Ecnomics, Engineering etc. all have their scientific languages for deepening their scientific discussions. Biology, even molecular biology, however, does not have a scientific language and its language is the "natural language". The issue "biological ontology" raised in this paper has deep incorrigible roots in the field where molecular biology is spreading. This paper examines two ontologies "Cyc" and "muKosmos" from the view point of applicability. The author's conclusion is that "To improve the current situation of non-unified and ambiguous vocabulary the only solution is to develop a core of commonly agreeable definitions and using those to implement user interfaces to and between databases." The reviewer is convinced of this conclusion theoretically through the paper. But mentally and by nature, the reviewer feels that it might be very difficult to reach a common core of agreeable "definitions" because the notion of "definition" (in the sense of Computer Science or Mathematics) does not exist in the minds of molecular biologists and the current "real molcular biologists" would not pay their attentions to the efforts for creating a "common core". A "wet" molecular biologist said to me with a sense of despise, "There are two kinds of biologists. One makes discovery and publishes papers. The other reads our papers and publishes something." However, "molecular biology ontology" is a hope which should provide a light into the mathematically dark era of molecular biology. This paper provides good criticisms and discussions arround this problem. - Recommendation: This is an excellent paper and can be accepted for ETAI as it is. |