OFFLINE

Welcome to Geisha, the Gallus Expression In Situ Hybridization Analysis
database. If you are accessing this page, you are likely one of our new
bioinformatics curators. Curators will be annotating and submitting entry
pages to the Geisha database. This database is the most comprehensive

repository for in situ expression data for the chick model organism, used by
scientists and teachers internationally. With your help, we can sustain and
expand this aim.

Please follow the instructions below until you become familiar with the duties of a
curator. If you have questions or notice a problem, please contact Dr. Diana
Darnell, the Geisha project manager at darnell@email.arizona.edu. And
again, welcome!

1. Log in
1.1 Open URL http://geisha.arizona.edu/geisha/curate.jsp

1.2 Scroll down to find a Geisha entry that is listed as ‘available’ in the status
column. Click “curate” to claim an entry (will change to ‘in process’), then
click ‘Edit’ to open the entry (or reopen one that you are working on).

1.3 In the middle of the page, choose your name in the pull down list
under ‘curator’.

1.4 Add today’s date to the date field (type or select) near the top.

1.5 Change library field to ‘student curator’

1.6 Notice the Geisha ID at the top of your entry. It is made up of the Gene
symbol for the gene you’ll be annotating, the first author’s last name and the
date of the publication separated by periods. Note: many papers contain
information on more than one gene, but each Geisha entry is for one gene
only. Please make sure you are curating data on the correct gene.

2. Reference
2.1 Look at the comments field on the entry page to find the reference. The first
author should match the name in the center of the Geisha ID at the top of the
entry. The year should also match. You will be finding this paper in PubMed
and adding its expression data for the gene identified at the beginning of the
Geisha ID into this entry.

2.2 Locate the paper in PubMed, the publications database for the National
Center for Biotechnology Information, NCBI. Open a new window or tab for
this URL:

http://www.ncbi.nlm.nih.gov/sites/entrez

2.3 In the search field, paste in the author list with names separated by commas,
and the first word of the title, also separated by a comma from the reference
you have been given. NCBI recognizes each term separated by a comma as

a unique search term.
For example: Tatsumi N, Miki R, Katsu K, Yokouchi Y, Neurturin

This should pull up a single abstract.
If it pulls up more than one abstract, pick the correct one.

2.4 To the right will be a button that says Full-Text Article, click it.
This will take you to the HTML version of the paper.

3. Probe Sequence
3.1 Locate the methods section of the manuscript, and the paragraph within
this that contains information on the in situ hybridization probes. Sometimes
(rarely) the information is contained in a table. Look specifically for
information on the probe associated with this gene name (the first term in the
Geisha ID at the top of the entry page you are curating).

3.2.1 IF the sequence of the probe is given in the paper, cut and paste it into the
complete cDNA template sequence box on the entry page. Then skip to
step 3.3

3.2.2 IF no specific probe sequence information is given (e.g., this probe was a
kind gift from Andrew Lumsden), then paste the following text into the public
comments box:
As insufficient information is provided in publication to verify exact
sequence used to synthesize probe, the sequence below was obtained
from NCBI (acc # ). Then skip to step 3.3

3.2.3 IF PCR primers are given, copy and paste the sentence containing
the information about the primers to the public comments
field. e.g., “The probe for XYZ was amplified by PCR using the
following primers: 5′-ATGCTCTTCTGCTCCTGTCGAG-3′ and 5′-
CTCGAGGGCCAGAACGACTCCTGGTGTGTT-3′.“ You will use this
information to extract the probe sequence from the NCBI sequence database.

3.2.4 If some other form of information about the probe is given (the probe
was produced from a subclone of the NM_2052101.1 cDNA containing
nucleotides 134-556), please cut out that information and put it in the
comments field AND use that information to extract the precise probe
sequence and add it to the box complete cDNA template sequence.

3.3 Open the URL for NCBI below, select Gene from the scroll down menu, and
type the gene symbol from the first term in your Geisha ID (e.g., GFRA2)
and gallus (Latin for chicken) into the search field.

http://www.ncbi.nlm.nih.gov/guide/

This should bring up the correct Entrez Gene page from the NCBI database.
The gene symbol will likely match the first term in your Geisha ID, but if not,
this could still be okay if there are multiple names for the gene (as frequently
happens). Always look for the full name on the Entrez Gene page and in your
paper. If they match, you’re good to go. If not, paste the full gene name from your
paper and gallus into the search box and try again.

3.4 On the Entrez Gene page in the second section called Genomic regions,
transcripts and products you can get to the longest known cDNA by clicking
on the blue link to the left of the genomic map (e.g., NM_2052101.1). In the
pop up box, choose FASTA (a format style).

3.5 In the Nucleotide window for the sequence of your gene’s cDNA, capture the
NCBI Reference Sequence name: (e.g., NM_205101.1) and paste it into the
box on the Entry page that you’re building called accession num. If you have
the text about ‘insufficient information…” in the public comments box, you
should also paste the Ref Seq at the end of that sentence where it says “(acc
# ) so that this will now read: (acc # NM_205101.1) for example. You should
also paste it into the field above the sequence fields that is called accession
num. Save.

3.6.1 IF the actual probe sequence is unknown, copy and paste the sequence of
this ref seq cDNA into the box labeled complete cDNA template sequence.
Save.

3.6.2 IF there were primers and the probe sequence is specifically known, cut
and paste the ref seq sequence into a word document (scratch paper) and
locate the sequence corresponding to the two primers used to amplify the
probe sequence. You should be able to do this by running FIND (option
F) and pasting the first primer sequence into the find window. Highlight
this text in the ref seq sequence. Next, find the reverse complement of the
second primer in the sequence. [Because the primer is given as all DNA
is written, 5’ to 3’, what is shown is the reverse complement of the single
strand of DNA sequence at which you are looking. Although there are
programs that will create the reverse complement of a sequence for you,
I find it easiest just to transcribe it myself into the find window. Starting at
the 3’ end of the sequence and reading backwards, type G for C and vice
versa, and A for T and vice versa.] 5′-CTCGAGGGCCAGAACGACTCCT
GGTGTGTT-3′ will become AACACACCAGGAGTCGTTCTGGCC (the
underlined sequence is a restriction site that has been added to the primer and
will not appear in the gene). Once you’ve identified the probe sequence that is
contained between these two primers, cut it out and paste it into the box called
complete cDNA template sequence. Save.

3.6.3 If the paper contained specific information about the probe, use that information
to determine the probe sequence to the best of your ability. If it gave nucleotide

numbers and a specific reference sequence name, paste the whole sequence in
question into word and use Tools: Word Count: Characters to help you select the
correct probe sequence. If it gives restriction enzyme sites as the ends of the
probe, make an effort to use the information appropriately. If you can’t figure
it out, move the text about the probe to the comments field and go back to step
3.2.2.

4. Photos
4.1 Scroll down to the first figure in the paper that contains an image of a chick
embryo labeled by in situ hybridization. You will likely have to read the figure
legends to determine the protocols used. Only normal or control embryos will
be added to the database at this time. We want the highest resolution images
from the paper that are available. If the figure window says in the HTML
version says High-resolution image, that’s the one we want. Alternately, the
PDF might be higher resolution than the HTML online if it is not high res. You
can open the PDF version by scrolling to the top of the page and selecting
PDF. Scroll down to the figure with the images you want, and increase the size
of the document to 200%.

4.2.1 Image capture for Mac: key command shift 4 and drag the window around
the panel you wish to capture.

4.2.2 Image capture for PC:
· Press the Print Screen key on your keyboard. It may be labeled [PrtScn].
· Open an image-editing program, such as Microsoft Paint.
· Go to the Edit menu and choose Paste.
· If prompted to enlarge the image, choose Yes.
· Optional: Use your image editor's crop tool to crop out unnecessary portions of
the screen shot.

4.3 Create a folder for images from this paper. Name it with the Geisha ID.

4 4 Go to the File Menu and choose Save As. Navigate to the folder where you
want to save the image.

4 5 Type a file name for the image. We use the Geisha ID + figure number
and panel ID (e.g., Wnt1b.Smith.2004.4A for figure 4, panel A) It takes a
minute but if someone else drags the photo onto their desktop from GEISHA,
it is already labeled with a citation of sorts for tracking purposes. Repeat steps
4.1-4.5 for all appropriate images in the paper.

4.6 Go back to the paper and read the legend for this panel. Usually it will give
the Hamburger Hamilton stage of the embryo (e.g., HH 16 or stage 7). This
is the number that goes in the stage box on the embryo page (see 4.8). If
there are different embryos at this stage for this probe in this paper, they will
get sequential numbers and be entered on different embryo pages. If there

are several photos of the same embryo (often whole mount and sections)
they will have the same embryo number and be entered on the same embryo
page. Copy all appropriate parts of the figure legend for this panel onto your
clipboard. For example, for a legend for Figure 3 with panels through K where
you are capturing panel C-E (one embryo) you would want to capture the
information in bold:

Figure 3. Expr ession patter ns of KLF s 2 in chicken embr yos between stages 1–27. D: Higher -
magnification view of the atr ioventr icular canal r egion of the embr yo in C. E: Section thr ough
the atr ioventr icular canal r egion of the hear t of the embr yo in C. H: Dorsal view of the embryo in
G. J: Dorsal view of the embryo in I. K: Section through the leg of the embryo shown in J; arrowheads
indicate faint labeling of the chondrogenic condensations.

4.7 In the Geisha entry, scroll down to the grey line labeled embryos and click
the add button on the right. This will take you to a page specific for the photos
of one embryo.

4.8 Enter the embryo number (see 4.6, default=1) and stage on this page and
click add to add the photo. This will take you to a new photo window.

4.9 In the photo window hit the browse button and select the folder and photo
you want to upload at this time. Open the file in the browse window to upload.

4.10 Paste the figure legend for this panel (these panels, one embryo) in the
figure caption box.

4.11 Select the anatomical regions shown expressing the gene in question
in this image from the ontology shown. This requires some familiarity with
anatomy. Sub-lists are hidden and can be opened by clicking the + next to a
main term. For example, Central Nervous system + will open a list containing
more detailed divisions of the CNS. If the words in the figure legend,
your anatomical knowledge or researching skills are insufficient to select
appropriate anatomical terms, write anatomy? in the comments field, and your
instructor or I’ll get back to you with suggestions during the revision phase.

4.12 Save (upper right hand corner) to return to the embryo page.

4.13.1 IF there are no other images of this embryo, save and return to entry
page.

4.13.2 IF there are other images of this embryo, hit add, and repeat 4.9-4.13 until
all images for this embryo have been added. Save.

4.14 Select the next embryo image to be uploaded and repeat steps 4.6-4.13
until all embryo images have been added and annotated. Save.

4.15 Once you’re back to the Geisha Entry page, check to be sure that the

embryo images appear in increasing stage order going down the page. If there
is a typo in the stage field, you can correct it by clicking the edit button for that
embryo entry to get back to the stage page. Save.

5. Informatics

5.1 Scroll to the bottom of the page and hit ssaha (bottom left). This will search
the ensembl database for sequence matches to the sequence you added
in step 3.6 and load the matches into a table called alignments. Because
your search term is a cDNA and the database you are searching is genomic
sequence, only the exons should match.

5.2 Check the table to make sure all of the exon matches are for the same
chromosome (conservation can lead to alignments with close family
members), and that the location on the chromosome for all of the exons
is reasonable (genes may span thousands or even tens of thousands of
nucleotides, but not tens of millions, however, small stretches of duplication
far away can be picked up by the alignment program). Delete inappropriate
exons. If you have a question about this judgment call, put it in the comments
or problems box. Save.

5.3 Near the bottom, click lookup ensembl gene in the lower left-hand corner.
This will match your alignments with the gene model that overlaps them. This
should bring up 1 ensembl gene whose gene symbol and/or description match
your Geisha ID or manuscript information for this gene. If you think they don’t,
write me a note in the comments field. Save. Yea! You’re done!

Optional:
5.3.1 To explore this process manually, copy the sequence from the complete
cDNA template sequence box at the top of the Geisha entry page onto your
clipboard. Open the ensembl URL at http://uswest.ensembl.org/index.html
and either choose chicken from the Search scroll down menu and your gene
name in the other search box and click through to the correct expression page.
Look at the proposed exon structure, chromosome and location. It should
match what the program found in its search. OR, click on the BLAST/BLAT
tab at the top and paste your clipboard sequence into the sequence paste
box. Select species (Gallus_gallus), dna database (LATESTGP) and search
tool BLAT, and RUN. This will show you the locations on the genome for all
of the matches with the best fits at the top of the list if you have it set to sort
by >score. Pressing the [C] in the left column on a line associated with the
appropriate chromosome and location will take you to a window with the gene
model (if there is one) mapped against the sequence you put in. Hopefully the
match will be a good one.

Email me with questions:
Jason Payne enyapja@hotmail.com

Diana Darnell darnell@email.arizona.edu