Through the genebank looking glass

As a contribution to the ongoing discussion of genebank database hell, we’re pleased to provide a platform for Michael Mackay, the person responsible for the bioinformatics projects at Bioversity International. At least until he gets a platform of his own …

Many thanks to Dirk van Enneking (and others) for providing some constructive input into suggesting what sort of information and functionality users of plant genetic resources (PGR) might actually want to see to help them locate and select accessions for their research and plant improvement activities. My understanding of the situation is comparable to Dirk’s.

• Is it a telling sign that there is only one response to my request for a profile of what germplasm users want from an information system in about four months?
• Is it such a boring and useless endeavor (to provide an online PGR information system) that it does not really matter what information and functionality is provided?
• Perhaps we have been socially engineered to accept mediocrity and now we are prepared to accept token systems?

In starting work on Global-ALIS (ALIS = Accession Level Information System), one component of the Global Information on Germplasm Accessions (GIGA) project that will build a single internet doorway to more comprehensive information on over 2 million accessions held in genebanks around the world, we are finding a lot of information to support Luigi’s original post Lost in genebank database hell and subsequent points raised in response — including Dirk’s contribution.

One thing we learn is that a lot of data has been accumulated but it is not all made available, possibly for reasons other than intellectual property issues. For example, several people involved in genebank PGR documentation have said that if no one uses the online systems (because they are not useful for the purpose of finding and selecting germplasm for research and plant improvement) then what is the sense in continually updating the data?

Another issue is the availability of characterization and evaluation data. At the moment it seems that only the GRIN system makes significant progress on this count.

A third issue is adding value to existing data. Luigi mentioned geo-referencing and implied spatial analysis as useful ways to better understand distribution and, perhaps, add value to existing data in term of quality and usefulness.

I am pleased to say that we are addressing these issues, and more, in the development of Global-ALIS. However, the eventual utility of this system (due to be deployed in early 2011) could reflect the input and suggestions of those who want to explore and utilize all that genetic variability buried deeply in the world’s genebanks.

We are about to build a Global-ALIS website where people will be able to make suggestions and comments on what genebank database heaven might be like, what features it would have and how people might use it. Dirk’s and everyone else’s positive comments are great stuff! We will certainly take these suggestions on board. The only caveat I need to make is that we cannot do everything at once, so please be prepared to prioritize your suggestions to help us address the issues that will have the most impact in the first instance.