-
Notifications
You must be signed in to change notification settings - Fork 1
Project proposal
Mark Fiers, Roy Storey, Marcus Davy
Lincoln University, Stuart Charters
Large amounts of data are continually generated in many fields of biological sciences. Meaningful ways to process and annotate this data are lagging. One good example is the sequencing and annotating of a genome. Generating the data is almost trivial but manual annotation is crucial for a good end result. Ideally, manual annotation is performed by the community using the genome sequence. However, there are few systems in place that allow collaborative annotation, moreover, they are almost always tailored specifically towards a single web site.
Recently, microblogging (for example: Twitter and Identica) has become ubiquitous as a uniform cross website method to comment on and discuss web resources. However, micro blogging systems do not provide a method to add semantics to a message and, hence, the ability to do any more meaningful searches than a keyword search on the generated data. If, for example, an article has a large number of tweets associated with it, the only way to find out what people thought is to read the messages, an onerous task that won't scale.
In this project we want to investigate if we can use microblogging to transfer semantic information by formulating a syntax to add semantics to a microblogging message, modeled after the W3C standard RDF. With this syntax we will build a proof of concept system to organize collaborative annotation of a genome sequence. The proof of concept system will consist of two parts. The first part will be web browser plugin that generates semantic microblogging messages from a Generic Genome Browser (GBrowse) site. GBrowse is a widely used system to display genome annotations. The browser plugin will add a number of interface elements to any page ranging from buttons that send a (syntactically correct) message that a certain annotation is correct or incorrect to a more elaborate interface that allows detailed annotation of a genomic element. The second part of the proof of concept will harvest the semantic messages from the micro blogging provider and aggregate them in a Semantic Mediawiki (SMW) site. Mediawiki is the wiki system that runs Wikipedia, Semantic Mediawiki is an extension to Mediawiki allowing storage and querying of semantic data. This SMW instance will allow GBrowse administrators, as well as end users, to search and assess annotations made.
If successful, this project will merge the ease of microblogging with the power of the semantic web. The openness of microblogging providers will allow for easy implementation of software to both create and harvest semantic microblogging messages. The method will be easily applicable to many different areas of science and does not have to be limited to collaborative genome annotation.
The project requires a student proficient with programming JavaScript, web applications and/or browser plugins. A working knowledge of RDF and semantic web applications is beneficial. The student will be required to interact with biological researchers to ensure usability of the application, hence the ability to clearly communicate with people outside of the students research field is very important.
In a world where more and more data is generated every day, any improvement in the way we can interact with this information is potentially important. More specifically, much of the research done in (bio-)sciences relies on the use of (often external) data sources. The quality of these databases is of paramount importance for the quality of the work downstream. This application will improve the ability of scientists to annotate and improve public databases.