Hi all,
Thanks for your comments and interest, both on and off the list. I am struggling to reply to everyone but will get there. A few comments to add to the discussison ongoing on this list:
Data quality/accuracy/reproducibility – This is definitely an issue. As Wolfram and others have mentioned we know that there is interlab variation in measurements and in many cases data are not normalized as they should be. There are a number of ways we can try to handle that, one option is to encourage users to provide methodological details (number and type standards, experimental error etc) and grade uploaded data based on the amount of information given. Datasets could b a assigned a 1, 2 or 3 star rating by database admins. This would also have the effect of grading labs, for example Lab X receives a 3 star IsoBank rating (before I receive an avalanche of hatemail, that is just an idea….). Ultimately, though I don’t see this as a flaw in the idea of combining isotope data. As a field, we should be producing reproducible, verifiable data. If IsoBank (or an other repository) can highlight instances in which this is not the case then that is something we need to face up to. Again, just my 2c
Other databases – Since embarking on this project I have learned about lots of great isotope databases typically built and curated by individual research groups or associated with specific sub-disciplines. One route IsoBank could take is to serve as an aggregator for these databases, whereby IsoBank automatically searchers for and includes data from waterisotopes.org (for example). As most of these databases are already open I don’t foresee much resistance to this idea. One great advantage is that this would allow us to compare, for example, water isotope ratios with tree ring data for multiple sites, or compare isotope ratios from modern or paleo samples from a single location. Fundamental to this working is creating a shares ontology of isotopes (basically a language which we all agree to use to describe our datapoints). If that is in place it become very easy to share, combine and integrate databases. This is one my next side projects and I’m eager to hear from others who are happy to share their isotope database schema.
Community interaction – As several colleagues have mentioned getting a critical mass of users will be key to this working. Journal requirements is certainly one way but although many journals now require data to be presented somewhere, I don’t think many specify a particular database. Therefore we as a community can decide where to place out isotope data. IsoBank could be one such repository but provided we have good linkages between repositories it shouldn’t really matter where the data are stored, provided they are stored in an accessible format.
Thanks for the continued discussion, it is great to hear other voices tackling these issues.
Best,
B
Dr. Brian Hayden
Science Manager
Stable Isotope in Nature Laboratory (SINLAB<http://www.unb.ca/research/institutes/cri/sinlab/>)
Canadian Rivers Institute
Department of Biology
University of New Brunswick
Fredericton, NB
E3B 5A3, Canada
Tel: +1 (506) 458-7148
Skype: brian.hayden.work
Twitter: @DrHaydo<https://twitter.com/DrHaydo>
Web: https://sites.google.com/site/haydenresearch/
Social Media Editor for The FSBI, an International Society for Fish Biology, find us on Twitter @TheFSBI<https://twitter.com/TheFSBI> and Facebook https://www.facebook.com/TheFSBI
From: Stable Isotope Geochemistry <[log in to unmask]> on behalf of Robert Buchwaldt <[log in to unmask]>
Reply-To: Stable Isotope Geochemistry <[log in to unmask]>
Date: Wednesday, March 22, 2017 at 11:41 AM
To: "[log in to unmask]" <[log in to unmask]>
Subject: Re: [ISOGEOCHEM] Why we need a centralized repository for isotopic data
There are several Databases around like http://www.earthchem.org/ so my question to the community if and how are the repositories different from the already existing and would the repository integrated into the already existing databases?
R.
On 3/21/2017 2:10 PM, Brian Hayden wrote:
Hi all,
The last few years have seen a groundswell of support for the idea of a centralized repository for stable isotope data. I have talked about this at the last two IsoEcol meetings and there are several researchers on this list actively working towards this goal. Last year, Jon Paul, Seth Newsome and others brought a few of us who are interested in this kind of thing to New Mexico to try put some structure on what a repository for stable isotope data should look like. I’m pleased to say that one of the outcomes of that meeting was an opinion article which has just been published in PNAS (http://www.pnas.org/content/114/12/2997.full.pdf)<http://www.pnas.org/content/114/12/2997.full.pdf%29>. There are many challenges to building a reliable and effective repository for stable isotope data but we hope that this paper at least serves as a ‘call to arms’ for the isotope community to achieve this goal. Anyone interested in contributing this this effort please keep in contact as there will hopefully be more developments on this front later in 2017.
All the best,
Brian
Dr. Brian Hayden
Science Manager
Stable Isotope in Nature Laboratory (SINLAB<http://www.unb.ca/research/institutes/cri/sinlab/>)
Canadian Rivers Institute
Department of Biology
University of New Brunswick
Fredericton, NB
E3B 5A3, Canada
Tel: +1 (506) 458-7148
Skype: brian.hayden.work
Twitter: @DrHaydo<https://twitter.com/DrHaydo>
Web: https://sites.google.com/site/haydenresearch/
Social Media Editor for The FSBI, an International Society for Fish Biology, find us on Twitter @TheFSBI<https://twitter.com/TheFSBI> and Facebook https://www.facebook.com/TheFSBI
|
