Hi Michel, Thanks for the questions and references. It would definitely help to clarify the perceived 'chasm'. In my experience of systems biology, understanding the control of different cellular processes on the behaviour of the cell/organism has been a central element. Approaches such as metabolic control analysis provides researchers with a framework to reach this goal, but what would be the analogue of this, if approached from the semantic side? Or is the goal different, to link models to knowledge frameworks and to make use of automated reasoning instead?
You also asked wether the approach or application create the chasm. To me the approach is clearly different. In the absence of quantitative models, I would say that the application also differs because its hard to imagine how one would predict the effects of quantitative changes in a system without such models. I should also add that the integration of data and models is an important issue at JWS Online and the SEEK, so I am glad to learn more. Best regards, Franco On 25 May 2012, at 4:50 PM, Michel Dumontier wrote: > Franco, > > On Thu, May 24, 2012 at 10:38 PM, Franco Du Preez <franco.dupr...@gmail.com> > wrote: > Hi all, > > As someone speaking purely from the kinetic modeling side, I have to admit > that there seems to be a quite a big chasm between 'traditional systems > biology' (if it has existed for a long enough period to be called that ;)) > and the semantic approach. > > in what what do you think there is a "chasm"? is it just in that the > approaches appear vastly different - one deals with values changing with > time, the other with truth values? Or is it that the applicability seems > unclear? If biomodels database is any indication, one can semantically > annotate the model entities with ontologies without much problem [1]. I and > others have shown how to use those ontologies to check the consistency of the > models [2]. More recently work [3], shows how we can integrate the results of > simulations in order to answer questions that spans beyond the original model > annotations. There are plenty more things that we can do now, particularly > in the context of enrichment analysis, association studies, rule mining, etc. > > m. > > [1] http://www.ncbi.nlm.nih.gov/sites/entrez/16333295 > [2] http://www.biomedcentral.com/1752-0509/5/124 > [3] > http://www.slideshare.net/micheldumontier/formal-representation-of-models-in-systems-biology > > > > This being said, I really looked forward to yesterdays session as it touched > on the interesting and practical application of model alignment, but alas, I > could not get my audio via the dial in. I guess many must have asked why were > not using skype, so I wont, but Mark's mail has prompted me to do more > digging and I finally got round to downloading a voip client that can dial > sip addresses (holding thumbs for next time). > > Cheers, > Franco du Preez > > On 24 May 2012, at 5:18 PM, Mark wrote: > > > Hi all SysBio'ers! > > > > I know it isn't really my place to be saying anything, but... when has that > > ever stopped me ;-) > > > > The last conference call was... odd?... and while Jun's extraordinary > > efforts to keep it moving forward were greatly appreciated (!! well done > > !!) I think it might be worth having a very open discussion about what our > > expectations are from this group, since it was a fairly small group and > > apparently with a wide range of experience and expertise. > > > > From what I could hear, there were four "tiers" of expertise in the group. > > Starting from the bottom: > > > > 1) People like me, who know nothing at all beyond that the SysBio/Modeling > > community have worked hard on putting together standards and technologies > > that are bearing fruit; and that I (as a mere potential user of the tech) > > need to become MUCH more aware of what they're doing in order to > > successfully pursue my own research interests. So... I'm the ultimate > > lurker on the call. > > > > 2) People like Erich, who know *a lot* about what's going on in the field > > (because this is their company's business!) but, as a vendor, isn't going > > to be the first one to speak in a call like this because it might come-off > > sounding like a sales-pitch. He's likely interested in both how the > > technology is evolving (to ensure their products stay current) as well as > > listening to the needs of the community (so that their products stay > > relevant), but I don't expect him to lead the discussion if for no other > > reason than he's simply too polite to "take-over" :-) > > > > 3) People like Jun, who has put in a lot of time learning what's out > > there, has a deep and genuine interest, and wants to discuss the pro's and > > con's of the various pieces at some level of detail with people who have at > > least tried to use it. (...but there weren't many! ...so she was speaking > > to herself most of the time...) > > > > 4) The full experts in the domain, most of whom were not able to make the > > call, unfortunately. And I don't say that in any way as an accusation, but > > rather, looking forward, I see a potential "boredom problem", which is what > > I think needs to be discussed. At least one of the domain experts who did > > attend, left the call mid-chat on the basis that it was "too simplistic" > > (exact quote from IRC)... so if we don't find a way to engage you, the > > experts, we might be in for some disappointing meetings! > > > > > > What I'd like to ask the SysBio community - especially category (4), since > > it seems to me that they are the critical ones to have on these calls, is: > > "what can WE (1), (2), (3) do to make these calls as useful to you as they > > will be to us?" I understand that you're probably already talking to each > > other, since this field is your "baby", and thus these calls have the > > potential to offer you little benefit beyond your existing email (etc.) > > chats! ...So... what can we do, as the broader-community, to provide > > value/feedback/etc. that would ensure we all - experts and noobs alike - > > get something useful out of this group and enjoy and value the hour that we > > spend together every couple of weeks? > > > > If I'm speaking out-of-turn, please flame me :-) I can take it! LOL! I > > just want to see this group succeed, and I am willing to stick my neck out > > to see if I can help! > > > > :-)===={ > > > > ^^^ > > my neck > > > > Best wishes all! > > > > Mark > > > > > > > > > -- > Michel Dumontier > Associate Professor of Bioinformatics, Carleton University > Chair, W3C Semantic Web for Health Care and the Life Sciences Interest Group > http://dumontierlab.com >
