Gramene's Rice environment ontology

[Pankaj Jaiswal]

Dear Graham,

My comments..

"McLaren, Graham" wrote:
> 
> Pankaj,
> I feel that your ontology is mixing up three important concepts. The
> environmental property, the scale of measurement and the actual value. You

I agree and we expect that having only the property will not add value to the
curation, that's why we want some kind of a scale in the ontology that fits
universally in most of the cases.


> need to concentrate on the property first e.g. temperature or location. You

Properties are there except for the farming practices like,  distance between
the plants, mixed farming (the type of crops/plants in the nearby environment)
and the weeds etc. But I believe most of them, that the researchers have
reported are already in.

I am getting rid of the location by GIS. but I am keeping the upland/lowland
etc.

> need to accommodate any scale the researcher wishes to use whether
> continuous or discrete, kg or lbs, and finally terms like Bhutan are values

Well this is a difficult situation. Right now I am in favor of having the
continuous measurement scale and value as a free text, because there is no set
limit/value. For the curation purpose, as we have seen most of the studies give
a comparative value which is easy to translate in  discrete values. This becomes
even important in case of QTL studies where in a given set of conditions and
strains the discrete values will remain same but the continuous values may
change significantly.
However to effectively translate it into the discrete values, I need to see some
of the data from INGER's evaluation work from nurseries, which is a standardized
practice. Is it possible for me to get their standardized charts to see what are
the environments that are applied/tested, the values and how they are scored.

> of property location and therefore do not have a place in the ontology.
> I think the crop modelers have done a lot of work on this try the DSSAT and
> IBSNAT sites.

I am getting rid of this.

Pankaj



> Graham
> 
> -----Original Message-----
> From: Pankaj Jaiswal [mailto:pj37 at cornell.edu]
> Sent: Monday, October 14, 2002 3:34 AM
> To: Michael Ashburner; Richard Bruskiewich; Graham Mclaren; Midori A
> Harris; Gramene at gramene.org
> Subject: Gramene's Rice environment ontology
> 
> Hi Everyone,
> 
> I thought MGED has already built up some kind of an environment ontology,
> but
> couldn't find one. So based on this presumption that there is none, I have
> compiled one for Gramene, which we are considering to adopt. The first
> version
> is available at
> 
> ftp://ftp.gramene.org/pub/gramene/ontology/environment/GR_env_ontology
> 
> There are some unresolved issues also and I need your feedback.
> 
> Pankaj
> _______________________________
> Unresolved issues:
> 
> like how to present the score, specifically with the abiotic environmental
> factors
> 
> A score may be defined on a
> 
> continuous scale such as for
>         growing duration in season/months/days
>         day length duration in number of Hrs
>         fertilizer amount applied in Kg/hectare
>         pesticide amount applied in ppm or gm/lt of solution
>         light spectral quality in wavelength
>         intensity in Lux etc.
>         temperature in Deg C/F
> 
> Discontinuous scale
> 
>         Temp. cold/warm/hot
>         application regime in terms of less (low)/more (high)/moderate
> (intermediate)/normal/optimal
> 
> On this pretext what we have come up with, is that we will for the sake of
> ontology use the
> discontinuous scale (embedded in ontology)
> 
> such as in case of temp
> 
> cold
> warm
> hot
> optimal
> normal
> 
>   %temperature regime ; GEO:0007123
>    %temperate ; GEO:0007126
>    %temperature cold ; GEO:0007124
>    %temperature normal ; GRE:0007164
>    %temperature optimal ; GRE:0007163
>    %termerature hot ; GEO:0007125
> 
> Same is for light intensity
> 
>         light intensity high/more
>         light intensity low/less
>         light intensity optimal
>         light intensity normal
> 
> and then if the author mentions a continuous scale, the specific info (like
> 13
> Deg C was the temp used in trait analyses for cold tolerance) goes in the
> comments section.
> I am still working on defining the discontinuous scales for each term (some
> are
> already in).
> 
> By doing this we expect, that it gets more generic. For example in case of
> different regions and
> cultivars or studies for that matter (specially a comparative analyses in
> QTL)
> these parameters on a finer level may vary, but they could still be called
> as
> either high/low/optimal/moderate/normal.
> 
> You may have something in mind on how we can resolve this issue? In my view,
> if
> we should bring the scoring attributes in the environment ontology,
> otherwise it
> is as good as a free text.
> 
> Lincoln Stein wrote:
> >
> > It looks very promising.  What are the major unresolved issues?
> >
> > Lincoln
> 
> ******************************************
> Pankaj Jaiswal, Ph.D.
> Postdoctoral Associate
> Dept. of Plant Breeding
> Cornell University
> Ithaca, NY-14853, USA
> 
> Tel:+1-607-255-3103 / Fax:+1-607-255-6683
> E mail: pj37 at cornell.edu
> http://www.gramene.org
> ******************************************

-- 

******************************************
Pankaj Jaiswal, Ph.D.                                   
Postdoctoral Associate
Dept. of Plant Breeding                             
Cornell University                                   
Ithaca, NY-14853, USA   

Tel:+1-607-255-3103 / Fax:+1-607-255-6683
E mail: pj37 at cornell.edu
http://www.gramene.org   
******************************************