Worm Breeder's Gazette 14(1): 20 (October 1, 1995)

These abstracts should not be cited in bibliographies. Material contained herein should be treated as personal communication and should be cited as such only with the consent of the author.


Sylvia. D. Martinelli1, Ian A. Hope2, Erik L. L. Sonnhammer1

1 The Sanger Centre, Cambridge CB10 1RQ
2 Dept.Biology, The University of Leeds LS2 9JT

Descriptions of 55 gene expression patterns (Class: Expr_pattern) have
been included in the worm release of ACEDB (4-8).  37 digitised images
of worms were also included, which can be fetched from within the
corresponding textual gene expression objects (Expr_pattern).

The pictures are currently only realised on UNIX systems with an
external viewer.  Any external viewer that can read jpeg files will do,
but we recommend using XV, which can be provided by us upon request. If
you are not using XV, you must replace the command "xv" in the file
wscripts/display_script with your own viewer command.  (Note that the
current system won't work on Macintosh, but this should be released

The pictures are accessed in two stages, leading from Expr_pattern text
objects to a Picture text object to the image itself. First select an
Expr_pattern which contains a Picture field. Click on the name of any
picture to call up the text window associated with a specific Picture;
clicking on the words Pick_me_to_call will produce a worm picture in an
XV window after some delay. Current pictures show worms with darker
stains representing lacZ activity. To remove the image, click with right
mouse button anywhere on the picture. This brings up the XV menu window
in which one can click on quit.

As well as the new class Expr_pattern, it was necessary to create Class
Picture.  Two other classes : Life_stage and Cell_group have been added
to maximise the utility of the data and give both temporal and spatial
qualities to gene expression.  The presence of both Cell and Cell_group
gives a more coherent structural hierarchy.  In consequence, some of the
so-called "Cells" in the Cell class have been moved into Cell_group
(such as vulvaD). Other Cell_groups represent tissue-types (such as
muscle) and organs (such as pharynx) but others are less well-defined.
An injection of more data would help with the refinement of the current
pragmatic rules for defining suitable objects in these classes.  The
classes Picture, Life_stage and Cell_group are not visible on the main
class menu but they can be examined by clicking with the left mouse
button on the triangle next to the words <In class.>, then selecting the
class wanted.  Small changes have been made to some classes (Paper,
Sequence, Clone, Protein, Locus, Cell) to permit cross-referencing with
the new text-based Expression patterns.

Those of you willing and able to submit expression data, please note
that the following data fields are additional to those listed when we
requested expression pattern data previously [wbg13.5p89] :-

Laboratory : as usual the two letter code if listed in ACEDB
Remark     : almost anything you like.
Picture    : a name for the picture file which needs to be in jpeg, gif
             or tiff format.
Life_stage : as specific as possible e.g L2 larva, postembryonic,
             post L3, adult male etc.
             If expression ceases during development, please indicate
Cell_group : a group of cells, either currently in database, or a new
             term embracing the cells affected e.g. tail ganglia,
             dorsal muscle.

             e.g. Cell_group e2
                  Cell e2DL
                  Cell e2DR
                  Cell e2V

             e.g. Cell_group rectal epithelium
                  Cell rect_VR
                  Cell rect_D
                  Cell rect_VL

This makes it clear to which Cell_group cells belong.

If you send in a picture you may wish write a brief summary of the most
cogent data from the Expression pattern under the heading DESCRIPTION.
This text will appear on the 'record card' called by clicking on the
Picture field from within an Expr_pattern window. Description should
include keywords, in particular the gene/clone/sequence in question and
the type of stain used, such as lacZ activity.

Further refinement will take time and requires constructive input from
the worm community : comments on the current presentation are sought,
also new data and pictures. Please send contributions directly to
sylvia@sanger.ac.uk in the future.