Notes from the biomass

The commencing day of the DILS’06 workshop started with reviews of several well known projects: Taverna, BioMoby (workflows), (webservices) and BioMart(data management/retrieval). If you require any such service, check them out, they are all well established projects with active communities (as most of the people reading this probably know).  The main subject of the day - workflows - definitely convinced me to explore the matter such as Taverna or Kepler again.

Noteworthy: Simon Mercer from Microsoft Research presented how Microsoft supports bioinformatics research in academia including projects small and large, e.g. Openwetware. IP generated in these projects remains with the academic scientists, Microsoft basically provides financial supports and receives insight into current research.

All in all, the workshop exceeded my expectations. The talks delivered much more than buzzwords despite - or may be because - being targeted at a small, experienced audience. The venue of the workshop, the Wellcome Trust Conference Center on the Genome Campus provides the right environment and infrastructure and the set up allowed for easy mixing with participants.
The field of data integration advances - not solving all problems as quickly as one would hope but I am convinced that webservices, data marts and workflows will hopefully replace Perl hacking in many places. Let’s see.

The auditorium is well populated. The European heat wave might contribute but I hear many rather happy remarks over coffee.

My personal highlight for today was Attempto Controlled English (ACE), an Controlled Natural Language approach provides both human descriptions and computer readable, presented by Tobias Kuhn from the University of Zürich.

(more…)

The workshop started on a strong keynote with Victor Markowitz providing overview, opinion and application to biological data integration. Many scientists both in Computer Science and Biology view the topic as a necessary evil at best and I might have been observed to support that one day or another but in this convincing keynote most acronyms were replaced by insight.

(more…)

The 3rd Internation Workshop on Data Integration in the Life Sciences 2006 (DILS’06), starting at the European Bioinformatics Institute in Hinxton, UK today is one of those workshops for the die hard bioinformaticians with the vision of structured, interoperable data sources in biology. Peter Norvig’s recent criticism of the semantic web, neatly summarized by Duncan at Nodalpoint probably affects the majority of the research that is performed by the attending scientists. While I do share many of Norvigs gloomy prospects on the lack of compliance for standards (etc.), giving up and let our current data babel continue feels like folding before picking up the cards. Just proposing data schemes by one consortium or another is certainly insufficient and we need to change the way we communicate and possibly conduct scientific research too - and changes are underway.
Back to the workshop: the talks won’t be breathtaking. Even well done presentations on ontologies, database schema and scientific workflows probably make excellent yet tough contributions to a Powerpoint Karaoke line up. Attending ten of them in a day can be tiring for someone with biological focus and one might get more from studying the projects website. Frankly, my main motivation is to meet and discover people to discuss and reflect on some concrete project ideas. Good talks would probably help but the workshop provides ample opportunity for discussions.

N.B. After re-reading Lem’s Futurological Congress recently, I wanted to reflect on its slander of science tourism in the next conference coverage. However, getting up at 3.45 a.m. and strolling through Cambridgeshire expelled any witticism on my part for now.

The latest issue of Nucleic Acid Research contains work that finds non-orthologous proteins in about one third of the COG database from the NCBI.
The work by Christophe Dessimoz et al. from the ETH Zürich entitled “Detecting non-orthology in the COGs database and other approaches grouping orthologs using genome-specific best hits” is low in voice and avoids clevering. Eugene Koonin, one of the inventors of the resource, admitted that the resource had some flaws in this respect.
The examples convince (e.g. for COG0508) as they follow up the screening for paralogs by a phylogenetic consensus tree.
However, the large number seems a bit worrying - I always thought that COGs would be rather too stringent and not contain many paralogs that could in principle be resolved. The finding that the majority of the wrongly included proteins have metabolic functions was likewise surprising.

The finding has only major implications if the majority of the non-orthologous can be shown to be functionally divergent, which I doubt. And can one use the procedure to provide a resource of the same quality as COG?

One of those days I am happy not to be a native English speaker. It must hurt so badly.

[Found here while looking for a tag line]

So, back blogging. I used the weekend to set up Notes from the biomass at this site and move from the old site, which will remain as is.
The set up is far from where I would want it to be, obviously. However, I want to cover the DILS workshop in Cambridge starting on Thursday, July 20th, hence this site must be running now.

More soon. Promised.