Bd[S] vs. Ser

FlyBase Project Members flybase at morgan.harvard.edu
Fri Jan 21 15:21:31 EST 1994




Dear Drosophilists,

The following conversation, summarized in the recent correspondence from
Alan Christensen has occurred over the last few weeks on the Drosphila 
bulletin board:

**************************************************************************
>Recently Yoshiaka Fuyama posted the following question:

>>>    The New Red Book says that Serrate (Ser) is a synonym of Bd[S]. But I
>>>  can't find any entry of Bd[S] in major stock lists though there are
>>>  numerous Ser. In a similar case, there are numerous bw[V1] but no Pm.
>>>    Is there sound reason for such inconsistency?
>
>To which Bill Gelbart replied, in part:
>
>>  You will undoubtedly find many such examples of inconsistencies between
>>  the New Red Book and stock lists.  These inconsistencies are generally
>>  historical or sociological, and evolve from the fact that different groups
>>  have been maintaining these different data files.  Common names in use for
>>  a long time (such as Ser) will only slowly be replaced in such independent
>>  stock lists by their newer proper synonyms (Bd[S]).
>
>Here's my two cents worth:
>Although I realize that Bd is the older name, and therefore takes precedence,
>I wonder if occasional exceptions shouldn't be made.
>In this particular case, I found the following occurrences in the stock lists
>on Flybase:
>                                 Number of stocks listed
>                                 _______________________
>Stock Center                     Bd      Bd[S]       Ser
>____________                     __      _____       ___
>Bloomington                       1         0         30
>Bowling Green                     1         0        177
>Umea                              0         0        119
>Ashburner                         0         0         29
>
>Although most of the occurrences of Ser are on TM3, which is kind of
>cheating, the point is that no one actually refers to Ser as Bd[S], and
>that very few alleles of Bd other than Ser even seem to exist any more.
>The addition of more lab stock lists would undoubtedly skew the distribution
>more heavily _towards_ Ser.
>
>...............................................................In the case of
>Bd[S], Serrate works.  Wouldn't it be better to just call it Serrate, and list
>Beaded as an allele of Serrate?  (See also Stephen Jay Gould's "Bully for
>Brontosaurus" for similar arguments in the Brontosaurus/Apatosaurus name
>debate).
>
>As a further argument for leaving Serrate alone are the fact that there are
>4 publications since 1990 whose titles refer to the Serrate locus, but none
>since 1949 that refer to Beaded (from the Flybase References section).
>Furthermore, both sequence entries in GenBank (DMSER and DROTMLPA) refer to the
>"serrate gene" and "serrate protein", and do not even list Beaded as a synonym.
>This means that anyone doing literature or sequence database searches must use
>the "incorrect" name to find anything. I revere the Red Book as much as anyone,
>and I agree that a consistent and predictable nomenclature is important,
>but I think Serrate is a good example of an exception that should be made.
>Such cases will probably be rare, but may be justified by the preponderance
>of usage, and by the fact that switching would cause more confusion than
>leaving well enough alone.

******************************************************************************


Alan clearly has a valid point.  The FlyBase group has discussed this
particular case and has come to the conclusion that on the basis of
the much more common usage of Ser, both in stock lists and in recent
publications, the community would be much better served by retaining 
the gene name Serrate and the gene abbreviation Ser, and by renaming
all Bd alleles as alleles of Ser.  This renaming will appear in the
next updating of the genes lists on the IUBio Gopher FlyBase files
(which will be called genes94).  It will take somewhat longer to 
implement changes to the FlyBase stock lists to make them consistent
with the updated genes and aberrations standards.  We hope to 
implement all of these changes by the summer or fall of 1994.

Some conflicts in names are inevitable, since allelism of two 
independently named entities is only realized after several years 
of research; this was exactly the case for Bd and Ser.  The other 
class of conflicts emerge when someone knowingly alters a gene name. 
(I confess remorsefully that I've been guilty of this trespass myself.)  
We wish to note that such conflicts must be resolved for the integrity
of the data in FlyBase, and that we will do so on a case by case basis.
As has been the tradition, unless there is a highly compelling 
justification to do otherwise in a specific case, we will continue 
to resolve conflicting names in favor of chronological precedence.

We recognize that it is vital to have a path that takes a user from 
an old name to its currently accepted form.  Users may well only know 
a gene by its old name.  We deal with this through the use of
synonyms.  We have already implemented a synonym table which is
available through the FlyBase gopher server.  Ultimately, when the full 
relational structure of FlyBase has been implemented, it should be 
possible to automate the process of connecting old names with their 
now-accepted synonyms through this synonym table.  We plan to build 
FlyBase in such a way that the user can query with any version of the 
name and still get connected to the right gene (or allele or aberration, 
etc.).



******************************************************************************

At another point in Alan Christensen's reply, he says:

>There are inconsistencies in the Red Book that don't seem to cause problems,
>most notably the peaceful coexistence of fa, Ax, Co, l(1)N, nd, spl, and N
>under the umbrella of Notch.  These have not been renamed N[fa], N[Ax], etc.
>This may not be the proper way to do it, but it seems to work.  

Actually, for systems such as Notch, we have rationalized their 
representation in the current genes93 directory of FlyBase.  You
will note that indeed fa has been renamed N[fa], Ax has been renamed
N[Ax], etc.  Since Notch is clearly a single gene by any reasonable
definition, we felt these changes were vital to developing a logical
relational database structure.


We appreciate the points that Drs. Fuyama and Christensen have raised, and
encourage members of the community to use the bionet.dros bulletin board to 
bring up other broad or focused issues of concern to the fly community, or 
to correspond with FlyBase directly (at flybase at morgan.harvard.edu) as they
prefer.

Sincerely,

Bill Gelbart  (for FlyBase)



More information about the Dros mailing list