Search Mailing List Archives

Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    


Waclaw Kusnierczyk waku at
Fri Jan 19 15:08:03 PST 2007

back to the recent changes in go, again.

'sensu ...' has been lately replaced with 'sensu ... research 
community'.  this change is supposed (i guess) to prevent confusion 
caused by the (still supported by go docs) convention of calling such 
terms 'species specific'.

i wonder what the reason to keep 'sensu' is.

consider this example (taken from the latest version of go available 
through amigo):

ADPG pyrophosphorylase complex
- ADPG pyrophosphorylase complex (sensu the Bacteria research community)
- ADPG pyrophosphorylase complex (sensu the Eukaryota research community)

the last two are, of course, 'as in but not restricted to ...'.  this 
means, the bacterial version is found in bacteria, but no claim is made 
that it is found only in bacteria.  analogously for the other one.  so, 
in principle, we may find out that some eukaryotes have this molecule as 
it is in bacteria, and conseqently annotate their genes/gene products 
with the bacterial version rather than with the eukaryotic.  though 
compliant with the (loose) semantics of 'sensu', it seems quite 
inelegant (if only lexically).

what is the difference between these two versions?  in the definitions 
it is stated that the 'bacterial' version is homo tetrameric, while the 
eukaryotic version is heterotetrameric.  the question is: why are 
reserch communities used as a criterion for differentiation, rather than 
what is essential to the distinction?  this could have been:

... complex
- heterotetrameric ... complex
- homotetrameric ... complex

note that in this case there would be no inelegance in annotating the 
homotetrameric version with an eukaryote gene product.  furthermore, 
such names are based on intrinsic (structural in this case) properties 
of what the terms represent, and are thus less likely subject to 
frequent updates.  in the case of 'sensu' names, if an eukaryote with a 
bacterial version of the molecule is discovered, we would have to either 
live with an inelegant annotation, or change the name.  (worse:  in this 
case, if the 'sensu bacteria ...' inclusion were removed to adapt for 
the eukarytic product, we would arrive at the parent term, thus would 
either have two homonymous terms, or would have to merge them into one.) 
  in the case of property- rather than community-based names, no such 
problem arises.

now, to indicate that the heterotetrameric and homotetrameric versions 
are (as far as we know) indeed found in distinct groups (in this case 
domains, taxonomically) of organisms, we can associate the term 
'heterotetrameric ...' with the taxon eukaryota, while the other with 
bacteria.  if we are sure that only eukaryota have the hetero version, 
we may mark the association as eukaryota-specificity.  if we are sure 
that all eukaryotes have this version, we may mark the association as 
validity (for example).  if we are only sure that some eukaryota have 
this version (but perhaps not all, and perhaps not only them), then we 
may mark th association as relevance (say).  these markings may later be 
changed, as our knwoledge (belief?) changes, or if the reality itself 
changes, and the association may no longer be of the type 'validity' or 
'specificity', or perhaps we may move 'validity' to some other taxon (if 
the hetero version does not appear in some eukaryotes, perhaps it 
appears in all plants, at least).  but the name of the terms, referring 
to the structure of the molecule, does not change.

it is also rather obvious that, to a human user as well as to an 
automated go term name parser, 'heterotetrameric' is a much more useful 
bit of information about what the term represents, as compared to 'sensu 
... community'.

there are perhaps cases in which it is not established what the 
difference between a 'sensu x community' version and 'sensu y community' 
version of a term (or what it represents) is (are there such cases 
really?), but in all other cases it seems reasonable to refer to the 
difference in what is represented rather than in who uses the term.

this discussion may not apply, however, to the cases where the same term 
  is used to represent different things rather than different versions 
of the same thing ('bud', even if not in go right now).  in this cases 
it may still be reasonable to use 'sensu ... research community' (and 
much more this than 'sensu ...'), at least temporarily.


Wacek Kusnierczyk

Department of Information and Computer Science (IDI)
Norwegian University of Science and Technology (NTNU)
Sem Saelandsv. 7-9
7027 Trondheim

tel.   0047 73591875
fax    0047 73594466

This message is from the GOFriends moderated mailing list.  A list of public
announcements and discussion of the Gene Ontology (GO) project.
Problems with the list?           E-mail: owner-gofriends at
Subscribing   send   "subscribe"   to   gofriends-request at
Unsubscribing send   "unsubscribe"  to  gofriends-request at

More information about the go-friends mailing list