, Last updated by jupp, on Wed, 12/07/2011 - 21:15

NEW: SWAT4LS tutorial http://www.e-lico.eu/populous_swat4ls_2011


Populous is an generic tool for building ontologies from simple spreadsheet like templates. The Populous approach is useful when a repeating ontology design pattern emerges that needs to be populated en-mass. The use of a simple interface, similar to that of a spreadsheet, means that the templates can be populated by users with little or no knowledge of ontology development. Once these teamplates are populated, Populous supports transforming the data into an OWL ontology using a expressive pattern language.

Spreadsheets are currently transformed into OWL/RDF using the Ontology Pre-Processing Language v2 (OPPL). OPPL 2 is a powerful scripting language for generating and manipulating OWL axioms. Populous provides a wizard like interface found in the "Tools" menu to map spreadsheet data to variables in OPPL patterns.

Populous is built on top of RightField. RightField can be used to create Excel spreadsheets that have ontology based restrictions on allowable values in selected cells. RightField spreadsheets allow scientists to annotate their data using standard terminology from ontologies rather than using free text annotations.

Populous and RightField are both open source cross platform Java applications. They use the Apache-POI for interacting with Microsoft documents and manipulating Excel spreadsheets.


The alpha release of the Populous extension (v0.9) is available here for download.



Populous requires Java 1.6.

1. Unzip the file

2. Windows user execute run.bat

3. Mac/Unix users execute run.command


Documentation is currently provided by a screen cast Demo of populous in action. There is a set of slides on NaturePrecedings from a recent presentation about Populous given at SWAT4LS 2010 here. The accompanying files used for the demo are provided in the example folder of the downloaded zip file.  If you are interested in this project and its development please contact me below for further details.

m name="allowscriptaccess" value="always" />
https://i.ytimg.com/vi/MQ_roJ7n2pc/hqdefault.jpg) !important;">



Simon Jupp (simon.jupp [at] manchester.ac.uk) and Robert Stevens (robert.stevens [at] manchester.ac.uk)



Ontologies are used to generate terminologies that describe the kinds (classes) of things (instances) we like to talk about within a particular domain. In the life sciences, for example, there are lots of kinds of things we like to talk about, and ontologies give us a mechanism to ensure we are talking about the same kinds of things. Standardising the way we annotate (or talk) about data makes it easier to integrate, process and analyse the data. In order for all of this to work we need to develop lots of ontologies to describe all the different kinds of thing we are interested in. Developing such ontologies is no mean undertaking, so we are constantly looking for new ways to reduce the ontology development bottleneck. One observation is that we often develop patterns to describe similar kinds of things, once these patterns have been identified, they can be left to domain experts to populate. Whilst ontology development environments provide support for template population, they often have steep learning curves, especially for users new to ontologies. We developed Populous as a light weight tool with a familiar spreadsheet style interface for domain experts to populate these ontology templates. The use of a transformation language, like OPPL, means we can separate the knowledge from the underlying ontological representation. This is particularly advantageous in situations where we want to radically change the modelling or offer different representations of the same data.



  • There are some issues in with reading Populous spreadsheets in certain versions of Microsoft Excel (especially Excel for MAC).
  • This current version of Populous is classed as Alpha, it has yet to be thoroughly tested on all platforms, so bugs are likely. Additionally, generating OPPL patterns can be difficult, especially debugging broken patterns. We are currently exploring better interfaces for the OPPL generation, but if you have any problems please contact us and we will be happy to help.