O'Really?

Last Friday, the UK National e-Science Centre in Edinburgh hosted a workhop, Semantic Web Applications and Tools for the Life Sciences (see SWAT4LS.org for the full details). Here are some incomplete and abbreviated notes from the workshop where there were some interesting people, paperware and software.

People and Paperware

• Nadia Anwar presented a paper on Semantic Data Integration for Francisella tularensi Proteomic and Genomic Data. This described experiences of developing a system for converting proteomics data in Excel to RDF, and what the benefits were.
• Heiko Dietze presented a paper on GoWeb: A semantic search engine for the life science web. Have a look at GoWeb query for the hexokinase enzyme for an example of the currently capability of GoWeb.
• Simon Jupp presented a paper on Knowledge Representation for Web Navigation which described why the Simple Knowledge Organisation System (SKOS) is sometimes a more suitable language for modelling knowledge than OWL and RDF – see also SKOS in the context of Semantic Web Deployment by Alistair Miles

70 people registered to attend SWAT4LS in total, many familiar names and faces, plus some new people I’ve never met before:

• Chemist James Procter, University of Dundee, enlightened me about DAS [1,2] – which has until now, been a bit of a Web service mystery to me. Sometimes it is much quicker to learn about something by asking an expert face-to-face, rather than reading the papers or playing with the software.
• Chemist Nico Adams, University of Cambridge. Only got a chance to briefly say hello, but maybe we’ll get a chance to talk more next time… Nico has written some more comprehensive notes on the workshop here and here.
• Christopher Baker, University of New Brunswick, one of the editors of the semweb-lifesci-book. On his way from Singapore to Canada (phew!)
• Icelandic bioinformatician Gudmundur Thorisson, University of Leicester. Nice to meet you and talk about genealogy in Iceland and DOI-goodness… [3]

Software

The workshop material had some adverts for publicationslist.org and a.nnotate.com which are potentially interesting tools that might be worth investigating. Publications list looks like a citeulike clone with the emphasis on claiming publications as your own, then publishing them as a list. A.nnotate.com looks like a handy tool for annotating pdf (and other) files… probably worth a look.

SWAT4LS finished with an entertaining keynote by Mark Wilkinson followed by a panel discussion hosted by Phil Lord titled: “If the semantic web is so good, how come most people use OBO for ontologies and PERL for data integration?” (Heh!) The simple answer was (by a fairly wide consensus) that the semantic web technology remains largely unproven. OBO is currently easier to use (!?) as well as having a larger number of more influential users (at least in the biomedical community). As Ben Good puts it (in a slightly different context), this makes semantic web technology either:

1. pitifully insignificant or
2. primed with the opportunity for dramatic growth…

We will probably have to wait a while longer to find out, maybe there is a third option? In the meantime, the SWAT4LS organisers (Albert Burger, Adrian Paschke, Paolo Romano and Andrea Splendiani) have published the workshop proceedings on http://www.ceur-ws.org (volume 435) with the best of these papers being published in a special issue by BMC Bioinformatics.

References

1. Andreas Prlic, Thomas Down, Eugene Kulesha, Robert Finn, Andreas Kahari and Tim Hubbard (2007). Integrating sequence and structural biology with DAS. BMC Bioinformatics 8 (1). doi:10.1186/1471-2105-8-333
2. Robin Dowell, Rodney Jokerst, Allen Day, Sean Eddy and Lincoln Stein (2001). The distributed annotation system. BMC Bioinformatics 2 (1). DOI:10.1186/1471-2105-2-7
3. Norman Paskin (2005). Digital object identifiers for scientific data. Data Science Journal 4, 12+. DOI:10.2481/dsj.4.12

[Creative Commons licensed picture of a statue of the scottish scientist James Clerk Maxwell in Edinburgh taken by David Farrer]