Standardized Term Names

[Alexander Diehl]

I'm concerned by recent changes performed to 
"standardize" term names.  In particular, I am 
dismayed by the altering of many "cell 
differentiation" terms to end with the term "cell 
differentiation" despite that fact that the word 
used for the cell type in question in many cases 
actually includes a suffix that means or implies 
"cell."

Thus, instead of "lymphocyte differentiation" we 
now have "lymphocyte cell differentiation," and 
instead of "adipocyte differentiation" we now 
have "adipocyte cell differentiation," et cetera. 
These modified term names are redundant phrases, 
in other words, bad English.

Furthermore, the terms in question are all is-a 
children of GO:0030154 cell differentiation, so 
logically they are already types of "cell 
differentiation" whether or not the phrase "cell 
differentiation" is part of the term name or not. 
This is a fundamental feature of the DAG 
structure of the GO, and there is no need to 
recapitulate the structure of the graph within 
term names themselves, particularly when the 
result is grammatically faulty.

Curiously, the explanation given for "y cell 
differentiation," "The designation 'cell 
differentiation' distinguishes these terms from 
other types of differentiation such as sexual 
differentiation" suggests somehow that naïve 
users of the GO might somehow miss the conceptual 
difference between go terms like GO:0007548 sex 
differentiation and GO:0045444 the former 
"adipocyte differentiation."  If so, I would 
suggest such users perhaps learn a little biology 
first, before tackling the GO, and then of course 
learn about the logical structure of the GO.

Also, anyone who bothers to read the definition 
of the terms in question would know immediately 
the sense of "differentiation" implied (assuming 
they hadn't first bothered to consider the term's 
placement in the DAG).  Isn't this why we put 
such stress on providing definitions for terms in 
the GO?

There are also evident faults in the 
implementation of this change.  As an 
immunologist, my first impulse might be to search 
on "lymphocyte differentiation" rather than 
"lymphocyte cell differentiation" yet the failure 
to retain the former term name as even a synonym 
means that I can't even find the term I am 
interested in.  I thought keeping former term 
names as synonyms was GO policy.

Plus, none of the regulation terms have been 
changed.  Hence we still have "regulation of 
lymphocyte differentiation" and "regulation of 
adipocyte differentiation," and so on.  This can 
only lead to even more confusion for users.

Furthermore, not all terms have actually been 
changed, such as GO:0042386 hemocyte 
differentiation (sensu Arthropoda) and GO:0042387 
plamatocyte differentiation.  Are there 
privileged terms in regard to this rule?

And finally, given the logic behind this rule, 
one might ask why GO:0030217 T-cell 
differentiation should not actually be "T-cell 
cell differentiation" given that "T-cell" is a 
type of cell and might well be written 
"T-lymphocyte" in an alternative GO universe. 
Similarly "B-cell cell differentiation."  I might 
enjoy that.

--Alex

-- 
Alexander Diehl, Ph.D.
Scientific Curator
Mouse Genome Informatics
The Jackson Laboratory
600 Main Street
Bar Harbor, ME  04609

email:  adiehl at informatics.jax.org
work:  +1 (207) 288-6427
fax:  +1 (207) 288-6131