Using ontologies to study cell transitions
© Fuellen et al.; licensee BioMed Central Ltd. 2013
Received: 21 January 2013
Accepted: 19 August 2013
Published: 8 October 2013
Understanding, modelling and influencing the transition between different states of cells, be it reprogramming of somatic cells to pluripotency or trans-differentiation between cells, is a hot topic in current biomedical and cell-biological research. Nevertheless, the large body of published knowledge in this area is underused, as most results are only represented in natural language, impeding their finding, comparison, aggregation, and usage. Scientific understanding of the complex molecular mechanisms underlying cell transitions could be improved by making essential pieces of knowledge available in a formal (and thus computable) manner.
We describe the outline of two ontologies for cell phenotypes and for cellular mechanisms which together enable the representation of data curated from the literature or obtained by bioinformatics analyses and thus for building a knowledge base on mechanisms involved in cellular reprogramming. In particular, we discuss how comprehensive ontologies of cell phenotypes and of changes in mechanisms can be designed using the entity-quality (EQ) model.
We show that the principles for building cellular ontologies published in this work allow deeper insights into the relations between the continuants (cell phenotypes) and the occurrents (cell mechanism changes) involved in cellular reprogramming, although implementation remains for future work. Further, our design principles lead to ontologies that allow the meaningful application of similarity searches in the spaces of cell phenotypes and of mechanisms, and, especially, of changes of mechanisms during cellular transitions.
The (artificial) induction of cell transitions has recently attracted a lot of attention. A cell phenotype (or cell type) can be defined by the cell’s repertoire of molecules and structural components at a certain time, together with the specific morphology and function they bring with them. A cell transition is a change in a cell that results in a new phenotype. For example, the phenotype of epithelial cells is distinct from the phenotype of fibroblasts. Programming of cells is the induction of a cell phenotype transition, e.g. from fibroblast to epithelial cell. Reprogramming is the artificially induced transition of a cell to a cell phenotype, which it (or its predecessor) had in the past. Potency can be defined as the disposition of a cell to transition into another cell phenotype; pluripotency is the ability of a cell to transition naturally into any of the cell phenotypes of an organism (where a transition is natural if it is not triggered by a technical intervention). Since Takahashi and Yamanaka described cell reprogramming of fibroblasts back to pluripotency (also known as generation of iPS, induced pluripotent stem cells) , hundreds of papers have dissected the reprogramming process and the cellular disposition of pluripotency at an ever-increasing resolution, reviewed in, e.g.,  and . This corpus is currently underused as there is no formal representation of the reported findings.
Several ontologies already exist in the domain of cell biology, such as the well-known Gene Ontology (GO)  and the cell type ontology (CL; cf. [5, 6]). Bard et al.  proposed formal definitions for CL classes, referring to properties of cells such as expressed proteins, activated biological processes, or phenotypic characteristics. Further cell-related knowledge projects include the Virtual Physiological Human project (http://www.vph-noe.eu/) that attempts to provide interoperability between different databases and tools related to human physiology and gene expression; the associated software Phenomeblast (code.google.com/p/phenomeblast) is an ontology-based tool for aligning and comparing phenotypes across species. However, many efforts in formal modelling of biological phenomena of organisms focus on anatomical features and only rarely address the cell level (cf. [7–10] and ). What is missing is a comprehensive tool to represent and to compare cellular phenotypes and their dynamics.
Results and discussion
Cell phenotypes and cell mechanisms
Most of the classes that are needed for the ontology of cell parts can be found in CELDA, the ontology developed by the CellFinder project (http://cellfinder.org/about/ontology/) , that itself integrates ontologies like the Cell Ontology (CL), the Cell Line Ontology (CLO), the Foundational Model of Anatomy (FMA), the Gene Ontology (GO) and Mouse Anatomy. Additional classes can be taken from the Cellular Phenotype Ontology  (CPO). For the ontology of cell mechanisms, we can re-use (portions of) the Interaction Network Ontology (http://bioportal.bioontology.org/ontologies/1515) and the GO subontology for biological processes (http://www.geneontology.org). The current GO (Biological Process), however, does not capture the hierarchical relationships described in Figure 2, connecting molecular events such as the interaction of Occludin and JAM to ultrastructural events such as the formation of a tight junction. Here, we need to explicitly encode the interconnections of molecular events and ultrastructural events. The ontology of cell parts (Figure 1) is designed to handle exactly the same challenge, on the level of the continuants. While the focus of the CPO is on phenotype abnormalities, we can still re-use it to provide distinct morphological and associated physiological phenotypes of cells and their components. Again, the hierarchical interconnections between the molecular entities and cellular parts (components, anatomical structures, cell types) need to be explicitly established, for example, between Occludin in tight junctions to the anatomy of specific cell types.
Independent continuants: Cells and their organelles as well as molecules are three-dimensional entities; they are present with all their spatial parts at every time of their existence.
Dependent continuants: Any property of a cell or a molecule, be it a quality or a disposition, also exists as a whole at every time of its existence. However, any such property is ontologically dependent on its particular bearer: It cannot exist without it.
Occurrents: Interactions, inhibitions, stimulations as well as transitions are temporally extended processes. They have temporal parts that occur at different times; hence they do not exist as a whole at any single point of time.
We can, for example, describe the state of a cell at a certain time by enumerating all of its parts and contents (independent continuants), or by enumerating all of its properties (dependent continuants), or by enumerating all the events going on at this time (occurrents), which could then be connected with parts and contents of the cell as their participants, e.g. with organelles or molecules. All of these categories are needed to integrate the data available: Cellular data describe continuants (like cellular components and dispositions) as well as occurrents, namely the molecular interactions (microscale mechanisms) going on in a cell at a certain time. While these continuants are covered by the phenotype ontology scheme, the interactions (microscale mechanisms) are covered by the mechanism ontology. Cell transition data describe occurrents, namely macroscale changes of microscale mechanisms. Within the EQ framework, we can describe such macroscale changes of microscale mechanisms by pairing terms for microscale mechanisms (as 'entities’) with specific change modifiers (as 'qualities’). In Figure 2 we illustrate this with one possible annotation pattern for a cell transition. In this annotation pattern, the entity term 'Interaction Occludin-JAM’ from the mechanism ontology is used as a subject term in combination with the qualifier 'up’ in order to express that in a certain time step the interaction between Occludin and JAM has been upregulated.
In our framework, a pluripotent cell can then be characterized by its expression data (about genes, proteins etc.), from which relevant microscale mechanisms can be inferred. A cell transition from one cell phenotype into another (e.g., of a fibroblast into a pluripotent cell) can be described by comparing the expression data of both cell phenotypes, which capture macroscale changes in microscale mechanisms. Such expression data include the start-up of the interactions between genes/proteins relevant for the induction of pluripotency. Such a start-up may happen because the cell starts to produce more instances of the molecule types participating in this type of interaction. In our framework, a pluripotent cell realizes dispositions for mechanisms relevant for pluripotency that may be described by a network of interactions. Further, a cell transition from fibroblast to pluripotent cell realizes dispositions for changes in mechanisms. After transition, the cell is characterized by the microscale mechanisms relevant for the pluripotent phenotype.
The use of the ontologies within the EQ framework
Our ontologies are designed to be used together with specific modifiers within the EQ framework. As shown on the right-hand side of Figure 1, the ontology of cell phenotypes can be used to collect annotations for cell phenotypes such as fibroblast, epithelial cell and pluripotent stem cell. We can set up annotation profiles of cells, consisting of sets of EQ pairs that describe them. For example, the profile of epithelial cells includes the information that the genes/proteins Occludin, the Junctional adhesion molecule (JAM), Claudin as well as tight junctions (TJs) are 'present’, and cell membranes are 'joined’. For this purpose, we can use a number of standardized modifiers like 'present’, 'absent’, 'up’, 'moderately up’, and 'down’, which can be integrated within an ontology like PATO . The quality terms used in a particular annotation profile are derived from the data being annotated, describing, e.g., a specific set of epithelial cells in a specific culture medium.
Both ontologies allow for annotation propagation. In , annotations for anatomical entities are propagated up a hierarchy of is_a and part_of relations, such that a parent receives all the annotations of its children. The rationale for this is the following. Every finger is part of a hand; hence any information about a finger is also information about some hand. Hence, whatever is explicitly annotated with “finger” is, implicitly, also about some hand. Annotation propagation makes these implicit annotations explicit via automated reasoning. In our domain, however, universal part_of relations are rare: As opposed to anatomical entities, molecules (like Occludin) and organelles are not usually restricted to one specific part of a cell or a specific cell, and process parts can belong to different process wholes. As a consequence of this, the mereological hierarchy cannot be used in the same way as in  for cell phenotypes and mechanism changes. As there is an implicit universal quantification over all instances of the first class in an assertion in an ontology description language like the Web Ontology Language (OWL) , we have to use has_part instead of part_of. In our example, a molecular entity like Occludin can belong to a range of cellular structures and phenotypes, while a certain cellular structure or phenotype has to possess certain molecular entities. Put in general terms, a cellular structure necessitates its essential molecular parts. That is, the whole determines the parts, and for this reason we need to use the has_part relation. The has_part relation is also appropriate for occurrents like cell mechanisms. This is because any initial temporal part of an event can happen without the event being completed. For example, not every S-phase needs to be part of a mitosis: the cell cycle can be disrupted, e.g. by the destruction of the cell that is about to divide, resulting in an S-phase that is not followed by a cell division at all. In contrast to this, every mitosis has an S-phase as one of its temporal parts. Again, we need to employ the has_part hierarchy, from whole processes to their necessary parts (e.g., from Network_of_mechanisms_relevant_for_TJ to the Interaction_Occludin_JAM). When employing annotation propagation, therefore, as a rule, a whole process will have a higher information content than its necessary parts.
The ontologies outlined above enable similarity searches across cell phenotypes and mechanism changes in analogy to . In particular, we wish to estimate the similarity of cell types and of cell transitions across time. In this setting, macroscale changes are processes happening from one state at a certain time to another state at a later time. What we call microscale mechanisms are activities around a certain time, i.e. activities that we suppose to happen at some (maybe small) interval around that time. Microscale mechanisms are typically described as undirected activities (interactions), while macroscale changes are of necessity directed to a certain end state. The EQ-syntax is used to build up annotation profiles for the cell types under scrutiny. If appropriate, they will be time-stamped in order to mark how much time has elapsed after a certain intervention (e.g., “two days after intervention X”). If a macroscale change is the transition from cell type A to cell type B, it can 'inherit’ the timestamps from the annotation profiles of A and B as its start and end time, respectively (see also Figure 3).
Similarity searches may then compare, e.g., EMT/MET and reprogramming data. In simplified terms, an MET (mesenchymal-epithelial transition)  consists in, first, the formation of adherens junctions (AJs) and, second, the formation of tight junctions (TJs). We represent the MET as the start-up of the microscale mechanisms relevant for an epithelial cell, which has as one of its parts TJ formation that is, in turn, represented as the start-up of the mechanisms relevant for a TJ. This is the inverse of an EMT (epithelial-mesenchymal transition, which happens in development, metastasis and fibrosis). ExprEssence and related tools ([20–23]) can be employed for generating annotations about mechanism changes relevant for a certain transition by means of high-throughput data analysis, and the more mechanisms are annotated, the better we can estimate how similar biological processes are. Ultimately, any set of cell transitions can be compared (using data coded in EQ syntax) with respect to the underlying mechanisms, demonstrating the power of our approach. Our ontology design principles thus enable a kind of BLAST search in the space of annotations (for mechanisms), with similar goals such as highlighting relationships (between mechanisms, based on basic mechanisms as building blocks), and eventually estimating their evolutionary history.
Clustering of cell phenotypes
We suggest that the improvement in similarity estimates afforded by our ontologies (see Figure 4) enables the plausible clustering of both cell phenotypes and cellular mechanisms. Then it should become possible to cluster cell phenotype and mechanism data sufficiently well to derive the clusters exemplified in Figure 5.
We outlined how to design ontologies that enable to (1) formally represent cell phenotypes and mechanism changes behind cell transitions such as (re-)programming, and to (2) develop algorithms exploiting this framework, including clustering and searching for similar cell phenotypes and mechanism changes. Both ontologies support manual curation of publication data, annotation propagation and information content measurement, as well as the inclusion of results from high-throughput data analysis.
Our use of EQ-syntax allows the systematic encoding of annotation profiles of cell phenotypes and mechanism changes. The terms for both types of entities are organized in hierarchies ranging from molecular to (ultra)structural to morphological entities. Annotation profiles can then be obtained using (1) data curation from publications or by (2) high-throughput data analysis. In ontological terms, bioinformatics tools such as ExprEssence can be used as an instrument for deriving mechanistic information from high-throughput data, turning information about continuants into information about occurrents by differential analysis. The starting point for expert curation, possibly supported by text mining, must be a set of carefully selected papers.
Given a rich annotated knowledge base, existing approaches for ontology-based similarity measurements  can be applied to the domains of cell phenotypes and cellular mechanism changes. This would yield two important functionalities: It allows clustering of cell phenotypes (and of mechanism changes) by similarity, providing important information for an operational definition of cell phenotypes, and it allows similarity search in the spaces of mechanism changes and of cell phenotypes.
To further refine and populate the ontologies, we currently explore the option to work together with collaborators in the DFG SPP 1356 (http://www.spp1356.de) on pluripotency and cellular reprogramming, and similar initiatives, and we are looking for further collaborations. The size of the final artifacts is obviously a function of time and efforts invested in their development. While the number of relevant entities is limited for cell anatomy and cell types (several thousands), it is very large and virtually unlimited for molecular entities.
To evaluate our approach, we intent to compare similarity search results based on high-throughput data analysis only to results based on employing the ontologies integrating high-throughput data, (ultra)structural data and morphological data, and further to compare both sets of results with the expectations of domain experts. We expect that in particular the relationships between molecular events (which may be derived from filtering high-throughput data) and ultrastructural events (curated from the literature) yield improvements for similarity searches (see Figure 4). To avoid a garbage-in, garbage-out scenario, the application domain must be strictly limited, e.g. to data describing reprogramming and EMT experiments, so that the input data can all be validated by domain experts. Ultimately we envision a community-based crowd-sourcing approach.
DFG support to AK and UL (AK 851/3-1, LE 1428/4-1), GF (FU 583/2-1, FU 583/2-2) and LJ (JA 1904/2-1) is gratefully acknowledged.
- Takahashi K, Yamanaka S: Induction of pluripotent stem cells from mouse embryonic and adult fibroblast cultures by defined factors. Cell. 2006, 126: 663-676. 10.1016/j.cell.2006.07.024.View ArticleGoogle Scholar
- Ho R, Chronis C, Plath K: Mechanistic insights into reprogramming to induced pluripotency. J Cell Physiol. 2011, 226: 868-878. 10.1002/jcp.22450.View ArticleGoogle Scholar
- Jaenisch R, Young R: Stem cells, the molecular circuitry of pluripotency and nuclear reprogramming. Cell. 2008, 132: 567-582. 10.1016/j.cell.2008.01.015.View ArticleGoogle Scholar
- Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig JT, et al: Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet. 2000, 25: 25-29. 10.1038/75556.View ArticleGoogle Scholar
- Bard J, Rhee SY, Ashburner M: An ontology for cell types. Genome Biol. 2005, 6: R21-10.1186/gb-2005-6-2-r21.View ArticleGoogle Scholar
- Meehan TF, Masci AM, Abdulla A, Cowell LG, Blake JA, Mungall CJ, Diehl AD: Logical development of the cell ontology. BMC Bioinforma. 2011, 12: 6-10.1186/1471-2105-12-6.View ArticleGoogle Scholar
- Gunsalus KC, Yueh WC, MacMenamin P, Piano F: RNAiDB and PhenoBlast: web tools for genome-wide phenotypic mapping projects. Nucleic Acids Res. 2004, 32: D406-D410. 10.1093/nar/gkh110.View ArticleGoogle Scholar
- Hoehndorf R, Oellrich A, Dumontier M, Kelso J, Rebholz-Schuhmann D, Herre H: Relations as patterns: bridging the gap between OBO and OWL. BMC Bioinforma. 2010, 11: 441-10.1186/1471-2105-11-441.View ArticleGoogle Scholar
- Hoehndorf R, Oellrich A, Rebholz-Schuhmann D: Interoperability between phenotype and anatomy ontologies. Bioinformatics. 2010, 26: 3112-3118. 10.1093/bioinformatics/btq578.View ArticleGoogle Scholar
- Hoehndorf R, Schofield PN, Gkoutos GV: PhenomeNET: a whole-phenome approach to disease gene discovery. Nucleic Acids Res. 2011, 39: e119-10.1093/nar/gkr538.View ArticleGoogle Scholar
- Hoehndorf R, Dumontier M, Oellrich A, Rebholz-Schuhmann D, Schofield PN, Gkoutos GV: Interoperability between biomedical ontologies through relation expansion, upper-level ontologies and automatic reasoning. PLoS One. 2011, 6: e22006-10.1371/journal.pone.0022006.View ArticleGoogle Scholar
- Washington NL, Haendel MA, Mungall CJ, Ashburner M, Westerfield M, Lewis SE: Linking human diseases to animal models using ontology-based phenotype annotation. PLoS Biol. 2009, 7: e1000247-10.1371/journal.pbio.1000247.View ArticleGoogle Scholar
- Gkoutos GV, Green EC, Mallon AM, Hancock JM, Davidson D: Using ontologies to describe mouse phenotypes. Genome Biol. 2005, 6: R8-10.1186/gb-2005-6-5-p8.View ArticleGoogle Scholar
- Mungall CJ, Gkoutos GV, Smith CL, Haendel MA, Lewis SE, Ashburner M: Integrating phenotype ontologies across multiple species. Genome Biol. 2010, 11: R2-10.1186/gb-2010-11-1-r2.View ArticleGoogle Scholar
- Seltmann S, Stachelscheid H, Damaschun A, Jansen L, Lekschas F, Fontaine JF, Nguyen-Dobinsky TN, Leser U, Kurtz A: CELDA-an ontology for the comprehensive representation of cells in complex systems. BMC Bioinforma. 2013, 14: 228-10.1186/1471-2105-14-228.View ArticleGoogle Scholar
- Hoehndorf R, Harris MA, Herre H, Rustici G, Gkoutos GV: Semantic integration of physiology phenotypes with an application to the Cellular Phenotype Ontology. Bioinformatics. 2012, 28: 1783-1789. 10.1093/bioinformatics/bts250.View ArticleGoogle Scholar
- Jansen L: Categories: the top level ontology. Applied ontology. Edited by: Munn K, Smith B. 2008, Ontos: Frankfurt, 173-196.Google Scholar
- OWL 2 Web Ontology Language Document Overview (Second Edition).http://www.w3.org/TR/owl2-overview/,
- Thiery JP, Sleeman JP: Complex networks orchestrate epithelial-mesenchymal transitions. Nat Rev Mol Cell Biol. 2006, 7: 131-142. 10.1038/nrm1835.View ArticleGoogle Scholar
- Ideker T, Ozier O, Schwikowski B, Siegel AF: Discovering regulatory and signalling circuits in molecular interaction networks. Bioinformatics. 2002, 18 (Suppl 1): S233-S240. 10.1093/bioinformatics/18.suppl_1.S233.View ArticleGoogle Scholar
- Guo Z, Wang L, Li Y, Gong X, Yao C, Ma W, Wang D, Zhu J, Zhang M, Yang D, et al: Edge-based scoring and searching method for identifying condition-responsive protein-protein interaction sub-network. Bioinformatics. 2007, 23: 2121-2128. 10.1093/bioinformatics/btm294.View ArticleGoogle Scholar
- Warsow G, Greber B, Falk SS, Harder C, Siatkowski M, Schordan S, Som A, Endlich N, Scholer H, Repsilber D, et al: ExprEssence–revealing the essence of differential experimental data in the context of an interaction/regulation net-work. BMC Syst Biol. 2010, 4: 164-10.1186/1752-0509-4-164.View ArticleGoogle Scholar
- Kim Y, Kim TK, Yoo J, You S, Lee I, Carlson G, Hood L, Choi S, Hwang D: Principal network analysis: identification of subnetworks representing major dynamics using gene expression data. Bioinformatics. 2011, 27: 391-398. 10.1093/bioinformatics/btq670.View ArticleGoogle Scholar
- Bhattacharya S, Zhang Q, Andersen ME: A deterministic map of Waddington's epigenetic landscape for cell fate specification. BMC Syst Biol. 2011, 5: 85-10.1186/1752-0509-5-85.View ArticleGoogle Scholar
- Huang S: The molecular and mathematical basis of Waddington's epigenetic landscape: a framework for post-Darwinian biology?. Bioessays. 2012, 34 (2): 149-157. 10.1002/bies.201100031.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.