Open Plant Data Format


#1

Hi there, I wondered if there is something like an open data format that can be used to exchange data about plants and plant growth (both environmental data like water temperature as well as plant data such as stem thickness or number of fruits).

Any feedback is welcome.


#2

To my knowledge, this is just getting started.
The environment data can be exchanged as json data (replication of the CouchDB), or as a csv extraction from the UI. This is currently available.
As for plant physiology, I don’t know of anything at this time. This is a difficult issue (I was a data architect at Monsanto for a number of years), and I would strongly suggest looking at OBO (Open Biological/Biomedical Ontology) for a start. I worked with it some in the past, and while it has some weaknesses, it is the best set of data definitions I know of; in particular the PO or plant ontology.


#3

thanks, this was exactly the kind of answer I was looking for. You seem to
be the person who actually has an overview and worked in this field, I am
glad my question found you. If we start working on plant ontologies and
data exchange protocols I will let that know below in the comments.


#4

Hi thanks for that, I was interested as well. I was also wondering if there is some sort of standard for applying a unique identifier to a particular plant variety? Or at least a unique identifier that is not 100’s megabases.


#5

To a particular genome (genetics)? No there is not, it tends to be an individual company giving it their own database identifier. Besides which, few companies of this sort really are open for sharing a lot of this information (hence why there is no standardization).
With a big seed company, that sequences (or partially sequences) thousands of seeds a week, it is difficult to have a broader standard. The only truly unique identifier would be the whole genome, but rarely is the whole genome sequenced; just key traits (that may vary by the company, or what research they are doing).


#6

Do you think that creating a standard unique identifier would be a good idea? Obviously it would obsolesce if and when fuller sequence data became routine for most farmers, but that might be a while.
I just noted that @wsnook was suggesting posting the seed company etc etc. I’ve been thinking about how to organize decentralized breeding projects, and it would certainly be a lot easier if the variety was identified uniquely instead of by the last person it was received from.