An Italian walks into a genebank… No, not the beginning of a tasteless joke. But still rather a funny story, so bear with me. The genebank is that of the International Rice Research Institute (IRRI), and the Italian visitor naturally starts discussing Italian rice with the (non-Italian) genebank manager. In particular, he tells him about this strange black rice they have in Italy, great tasting and great for the health, an old traditional Italian variety, dating back to time immemorial. Venere, it is called. Does the IRRI genebank, the largest rice genebank in the world after all, perhaps have it?
So the IRRI genebank manager, who likes that sort of challenge, does a bit of googling, and a bit of trawling of the various databases at his command. And what should he discover, but that the old traditional Italian variety actually traces back to a donation made by IRRI 22 years back. Yes, records agree that in 1991 IRRI sent a sample of a black rice from Indonesia (IRGC 17863, local name “Ketan Gubat”), collected in 1972 in a place called Yogyakarta, to W.X. Ren at the Italian seed company Sardo Piemontese Sementi. Venere is a descendant of Ketan Gubat.
Helping develop a new market in speciality rice has not traditionally been counted among IRRI’s impacts. There’s a first time for everything, I suppose. But there’s a bigger point here. How do you monitor this kind of impact pathway in a systematic way? Continuously chase literature on the use of all the hundreds of accessions a genebank sends out? Unlikely. ((Or is it? Could it be automated, perhaps?)) No, what you really need is a curious visitor who knows a particular variety walking into your genebank.
Mamma mia! I don’t remember sending that rice . . .
And that, I suppose, is why we need databases.
Great story. I’m off to IRRI soon, to discuss Darwinian Agriculture.
As DNA sequencing gets cheaper (see “Son of Moore’s Law,” by Dawkins), identifying the source of germplasm could indeed be automated. Identifying the source of ideas is trickier. If I cross two plants, each leaves a clear fingerprint in the DNA. But if I combine two or more ideas, even I may not remember where they came from.
Tracking ideas may not be far off. DNA and human language both being codes, and new ideas typically giving rise to new language combinations, if not new words, it is at least theoretically possible watch ideas combine. The concept of relative entropy, or surprisal, which measures the relative predictability of certain words or strings based on previous usage, is already used to determine unknown authorship (as in this paper: http://ww2.cs.mu.oz.au/~jz/fulltext/airs06.pdf), it may only be a step more complex to track relative entropies in an academic discourse for example, and watch the entropy spike when a new idea is introduced, then decline as the idea is adopted and the language in the field becomes more predictable. This would be roughly analogous to the process of natural selection, when a mutation is introduced, then either copied or discarded.
talking about ideas and how they spread… Here is one that may or may not be mine, but it soon will also be yours: Lets sequence all that germplasm for food and agriculture in the public domain and finally put numbers to the access and benefit sharing mechanisms called for in the ITPGRFA. Looks to me as if a big share of the revenue from Venere should go to Indonesia!
Good idea. Maybe the Indonesians will use the cash to plant more palm-oil plantations.