[Go] precomposition = denormalization?

[Jim Hu]

Chris,

With all the discussion about column 16 and pre vs post composition.   
Isn't post-composition just the equivalent of database normalization,  
and don't the usual arguments for normalization apply?

I find myself wondering if specific GO terms will need some field that  
specifies the allowed cross-products. For example, I'm not sure it  
would ever make sense to do a cross product to an anatomy term for a  
binding molecular function term, and it wouldn't make sense to do a  
ChEBI cross product for a process localization.  I'm thinking that  
cross-product allowed ontologies would propagate to child terms.

In other words, the precomposition would be to what ontologies/data  
sources provide allowed cross products, rather than to the cross  
products themselves. That would simplify building tools to post- 
compose the terms at annotation time in a way that is  
indistinguishable to the end user from precomposition.  That would  
address the UI issues, I think, and moot much of the current  
discussion, no?

Wish I could be in Eugene... but others may be just as happy that I  
won't be!

Jim
=====================================
Jim Hu
Associate Professor
Dept. of Biochemistry and Biophysics
2128 TAMU
Texas A&M Univ.
College Station, TX 77843-2128
979-862-4054


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://fafner.stanford.edu/pipermail/go/attachments/20090324/90388ba0/attachment-0001.html>