=Paper=
{{Paper
|id=Vol-2285/ICBO_2018_paper_62
|storemode=property
|title=Reasoning over Anatomical Homology in the Phenoscape KB
|pdfUrl=https://ceur-ws.org/Vol-2285/ICBO_2018_paper_62.pdf
|volume=Vol-2285
|authors=Paula Mabee,James Balhoff,Wasila Dahdul,Hilmar Lapp,Christopher Mungall,Todd Vision
|dblpUrl=https://dblp.org/rec/conf/icbo/MabeeBDLMV18
}}
==Reasoning over Anatomical Homology in the Phenoscape KB==
https://ceur-ws.org/Vol-2285/ICBO_2018_paper_62.pdf
Proceedings of the 9th International Conference on Biological Ontology (ICBO 2018), Corvallis, Oregon, USA 1
Reasoning over anatomical homology in the
Phenoscape KB
Paula M. Mabee James P. Balhoff
Department of Biology Renaissance Computing Institute
University of South Dakota University of North Carolina at Chapel Hill
Vermillion, SD, USA Chapel Hill, NC, USA
Wasila M. Dahdul Hilmar Lapp
Department of Biology Center for Genomic and Computational Biology
University of South Dakota Duke University
Vermillion, SD, USA Durham, NC, USA
Christopher J. Mungall Todd J. Vision
Environmental Genomics and Systems Biology Department of Biology
Lawrence Berkeley National Laboratory University of North Carolina at Chapel Hill
Berkeley, CA, USA Chapel Hill, NC, USA
Abstract— The Phenoscape project (www.phenoscape.org) genes, we incorporated homology reasoning in the Phenoscape
has semantically annotated the features of species from the Knowledgebase (KB) (kb.phenoscape.org). One of the
comparative literature, enabling links between candidate genes difficulties in rendering homology knowledge amenable to
and novel species phenotypes for which they might be reasoning is that statements of homology are hypotheses, and
responsible. To enable discovery of homologous phenotypes and in some cases homology assertions regarding the same
associated genes, we incorporated machine-reasoning with anatomical structures can be in conflict. Thus, we represent
knowledge about homology into the Phenoscape Knowledgebase homology assertions separate from a core anatomy ontology as
(KB). We show that with homology reasoning enabled, the results
annotations in spreadsheet form with evidence and attribution.
of database queries can be expanded to incorporate shared
The annotations are transformed into OWL axioms according
evolutionary history. We discuss the challenges in developing a
logical model of homology assertions and implications for to a model with the desired entailments, and a user can choose
database queries, as well as theoretical entailment and practical whether or not to include hypotheses of homology in
performance tradeoffs between alternative models. reasoning. We explore the ramifications of different logical
models of homology and use a series of competency questions
Keywords—homology; anatomy ontology; phenotypes; to evaluate the performance of each model.
reasoning; evolution
II. ANNOTATION OF HOMOLOGY ASSERTIONS
I. INTRODUCTION Homology assertions for both historical and serial
The enormous volume of biological data that has become homology of vertebrate skeletal elements were extracted from
available to researchers has brought with it a rapidly expanding the comparative literature for teleost fishes and early
taxonomic range represented by the data. Because different sarcopterygians [3], and from the developmental genetic
taxa can possess similar features due to shared ancestry, the literature. We constructed these assertions using anatomy terms
incorporation of homology in connecting, aggregating, and from the Uberon anatomy ontology [4] and taxon terms from
analyzing data has become increasingly important. For the Vertebrate Taxonomy Ontology [5], resulting in a total of
example, without explicit incorporation of homology, the 98 homology assertions pertaining to skeletal anatomy.
results of user queries for species phenotypes or candidate Attribution for each homology statement was recorded, and the
genes may be incomplete. The Phenoscape Project [1, 2] has type of evidence (e.g., positional, developmental) provided by
semantically annotated the features of species from the the author supporting or contradicting the homology assertion
comparative literature, enabling links between novel species was annotated with terms from the Evidence and Conclusion
phenotypes and candidate genes that may underlie them. To Ontology [6]. The most common type of evidence for or
enable discovery of homologous phenotypes and associated against homology cited by authors in the collection of
Funding provided by National Science Foundation ABI Innovation
collaborative grants (1661529, 1661356, 1661456, 1661516) and an ABI
Development grant (1062542).
ICBO 2018 August 7-10, 2018 1
Proceedings of the 9th International Conference on Biological Ontology (ICBO 2018), Corvallis, Oregon, USA 2
homology assertions were based on development (27 homology reasoning incorporated, computational tools can
statements), followed by morphological similarity (26 now access the results of reasoning across evolutionary history.
statements), position (20 statements), and gene expression Although the model we select and implement in the KB
(14 homology statements). Some author statements (5) cited satisfies basic reasoning, we expect that it can and will be
evidence traceable to a different publication, whereas some (6 optimized for different purposes, and as computational
statements) did not cite traceable evidence. The collection of methods to represent uncertainty evolve.
homology assertions was incorporated in the Phenoscape KB,
which currently contains over 600,000 annotated phenotypes ACKNOWLEDGMENT
for vertebrate taxa from 185 comparative morphological
studies. We are grateful to the many collaborators who have
contributed data and expertise to the Phenoscape Project.
III. HOMOLOGY REASONING MODELS
REFERENCES
We have taken an exploratory approach toward resolving
[1] R. C. Edmunds, B. Su, J. P. Balhoff, B. F. Eames, W M. Dahdul, H.
the most effective way to enable machine reasoning on Lapp, et al. (2016) Phenoscape: Identifying Candidate Genes for
historical and serial homology across anatomical structures. Evolutionary Phenotypes. Mol Biol Evol. 33: 13–24.
Specifically, we explored the reasoning ramifications of two [2] P. M. Mabee, J. P. Balhoff, W. M. Dahdul, H. Lapp, P. E. Midford, T. J.
OWL models of homology that we have developed. In the first Vision, et al. (2012). 500,000 fish phenotypes: The new informatics
model, classes of homologous entities are represented using landscape of evolutionary and developmental skeletal biology. J Applied
reciprocal existential property restrictions. In the second Ichthy 28(3):300-305.
model, an OWL individual is introduced that represents the [3] T. A. Dececchi, J. P. Balhoff, H. Lapp, P. M. Mabee. (2015). Toward
synthesizing our knowledge of morphology: using ontologies and
ancestral structure from which all instances of two classes of machine reasoning to extract presence/absence evolutionary phenotypes
homologous structures are descended. Using the collection of across studies. Syst Biol 64(6): 936-952.
homology assertions and a sample of fin/limb phenotypes from [4] M. A. Haendel, J. P. Balhoff, F. B. Bastian, D. C. Blackburn, J. A.
the KB, we evaluated each model against the expected Blake, A. Comte, et al. (2014). Uberon: Unification of multi-species
outcomes for a set of queries formulated as competency vertebrate anatomy ontologies for comparative biology. J Biomed
questions relevant to research in developmental biology, Semantics. 5:21
comparative anatomy, and evolution. We discuss these results [5] P. E. Midford, T. A. Dececchi, J. P. Balhoff, W. M. Dahdul, N. Ibrahim,
and the implementation of homology reasoning in the KB. H. Lapp, et al. (2013). The Vertebrate Taxonomy Ontology: A
framework for reasoning across model organism and species
phenotypes. J Biomed Semantics. 4:34.
IV. CONCLUSIONS [6] M. C. Chibucos, C. J. Mungall, R. Balakrishnan, K. R. Christie, R. P.
Huntley, O. White, et al. (2014). Standardized description of scientific
We have enabled homology reasoning in the Phenoscape evidence using the Evidence Ontology (ECO). Database. 2014:bau075
KB, where it allows discovery of homologous structures. With
ICBO 2018 August 7-10, 2018 2