From: Alan R. <ala...@gm...> - 2008-06-24 15:57:09
|
On Jun 24, 2008, at 9:23 AM, Fostel, Jennifer (NIH/NIEHS) [C] wrote: > thanks. yes, the protocols are "the same" except for the parameter > values, as you have written. > > we were hoping it would be possible to list the paramters, for example > protocol parameter = dosage, value = 20 g > but possibly this is a matter for the implementation? > fishing them out of the code example you provide will befuddle the > user OK. Well the first step was to determine whether the interpretation is correct. At least we know that if we do this sort of abbreviation, what it is an abbreviation for. I'm also not sure whether this belongs directly in the representation or whether it is an implementation issue. Good topic to discuss at the meeting. -Alan > > to comment on one of your thoughts, we never know the mass exactly, > just > a measurement of it :-) > how else would we know the mass? The thing is that in some cases, we have quality instances that are full specified, like "symmetric", but in this case, we wouldn't, in some sense. -Alan > > ...jennifer > > -----Original Message----- > From: Alan Ruttenberg [mailto:ala...@gm...] > Sent: Saturday, June 14, 2008 7:44 AM > To: obi...@li...; Fostel, Jennifer (NIH/ > NIEHS) > [C] > Subject: Re: [Obi-plan-branch] protocol parameters > > I'll give it a first pass, though it may not be completely baked. > > On Jun 13, 2008, at 1:08 PM, Fostel, Jennifer (NIH/NIEHS) [C] wrote: >> Hello, all! >> >> we have several very similar protocols which differ at the parameter >> specification. we were thinking of these as the "same" >> protocol with different parameters. >> > One can see the conflict even in your phrasing. Are they very similar? > or the same? I think we should reserve "the same" to mean the same, > and > instead consider this a standard situation of having a class and > subclasses. (assuming protocol is a class, but that's a separate > discussion). > > So: >> for example, dosing-a-rat protocols within a study would be identical >> except for the dose level can we use OBI to annotate this? >> > For the sake of discussion, suppose that the drug is aspirin > measured by > mass, and let's not get into the question of whether we know the mass > exactly or just have a measurement of it. I'll write "mass of n grams" > as the name of a mass quality for this purpose. It probably means the > instance of mass quality that was measured to be n grams. > > I'm writing this at the class level, since I'm assuming that that's > what > you requested, though it may be appropriate to consider this at the > instance level. > > portion_of_aspirin: acetylsalicylic acid and has_role some drug_role > > Dosing-a-rat-with-aspirin: material transformation and has_input some > rat and has_input some portion_of_aspirin and has_output some (rat and > has_role treated_role) and (exactly 1 has_input (has_role some > drug_role)) > > Dosing-a-rat at 10mg: Dosing-a-rat and has_input some (aspirin and > (has_quality some (mass of 10mg))) Dosing-a-rat at 20mg: Dosing-a- > rat > and has_input some (aspirin and (has_quality some (mass of > 20mg))) ... > > This doesn't preclude the description of a larger protocol with > multiple > dosings. > > -Alan > > > >> >> (the protocol actually has about a dozen parameters, but to make it >> straightforward i list only one) >> >> thanks! >> >> ...jennifer >> >> >> Jennifer Fostel, Ph.D. >> CEBS Scientific Administrator >> SRA International >> >> Laboratory of Respiratory Biology >> NIEHS, NIH >> PO Box 12233 Mail Drop F1-05 >> 111 Alexander Drive >> Research Triangle Park NC 27709-2233 >> >> phone 919 541 5055 >> fax 919 541 1460 >> >> --------------------------------------------------------------------- >> - >> --- >> Check out the new SourceForge.net Marketplace. >> It's the best place to buy or sell services for just about anything >> Open Source. >> http://sourceforge.net/services/buy/ >> index.php_______________________________________________ >> Obi-plan-branch mailing list >> Obi...@li... >> https://lists.sourceforge.net/lists/listinfo/obi-plan-branch > |