Antibody:X

Unique antibody target
Jakob: We need to make sure that antibody descriptions contain a single unique target. Tubulin for example could be one of many protein - tubulin alpha, beta, gamma,.. The UniProt ID is unique but not human readable. Standardised gene names with species would be another way. The UniProt variety of this would be TBG1_HUMAN in our example, which I think is much more informative than P23258. http://www.uniprot.org/uniprot/TBG1_HUMAN also works as a URL for UniProt but the 1st part TBG1 is unfortunately not the standardised gene name TUBG1.

Maybe we could do some content verification in the background using either NCBI (http://www.ncbi.nlm.nih.gov/sites/entrez?cmd=search&db=protein&term=P23258) or UniProt (protein pages contain code like this: P23258|TBG1_HUMAN Tubulin gamma-1 chain OS=Homo sapiens GN=TUBG1").

Type field hide/remove
Jakob: Antibody-cat2-name is not very important information for the average reader. Can we hide/delete it? What is it for exactly? I wanted to ask before deleting it myself prematurely.
 * Jakob: removed

Uniprot links
Jakob: Is it possible to generate a link from the UniProt ID similar to PMID 1234567? Something like Uniprot P23258. As far as I understand UniProt currently has the following URL structure: http://www.uniprot.org/uniprot/P23258
 * Bill Flanagan (OpenWetWare.org at MIT) 09:55, 22 June 2009 (EDT)Done. Added uniprotid field
 * 15:13, 22 June 2009 (EDT)I've clearly mislabeled this. We should find a way to make display conditional on setting the target as a protein.
 * I'm not currently testing for a valid link but it may be useful: don't include it if it will throw an error. Good idea for a simple extension to only allow a link if it works. Later.
 * Jakob Great. Although a majority of antibodies is against proteins, this field will not be necessary for some antibodies. A link check would be great.

Provider field
Jakob: I'm wondering whether antibody provider should be an infobox field. Nowadays, most antibody are purchased from companies instead of generated in the lab. In other words, most antibodies will have 1+ commercial provider that could be linked to here or in the main page.
 * Bill Flanagan (OpenWetWare.org at MIT) 09:55, 22 June 2009 (EDT)Done. Added provider and link to example
 * Should we make this always generate link to oww page?
 * Do not want people putting url's here.
 * make oww provider page instead. Link can go there.
 * Bill Flanagan (OpenWetWare.org at MIT) 15:13, 22 June 2009 (EDT)This is how it now is implemented. Put in the name of the provider. The info is formatted as an OWW link. Points to Provider:Antibody/Provider-Name where Provider-Name is the provider entered
 * Jakob: We need a direct link to the antibody page of the provider somewhere - if not in the infobox, then in the main page. A link to an OWW page describing the provider is not really necessary. You want to avoid direct links because..? Since there may be multiple provider that need explaining, we may want to move this section into the main body altogether, even have a default section for provider description/discussion integrated into the automatic antibody page creation utility.
 * Bill Flanagan (OpenWetWare.org at MIT) 15:52, 25 June 2009 (EDT) OK. I'll add it and place it in the body. No problem. The link will contain the company name. I think we should assume there may be more than one provider. It will look like this:


 * 1) add provider-url to the list of fields.
 * 2) place first provider name in the infobox.
 * 3) populate the first provider entry in the text.
 * 4) prepend the section with instructions on how to add another one.

"Enter the provider name here" (no quotes!)

 * Url: ["Enter the provider company url here"(no quotes!) "Enter the provider name name here"(no quotes!)]

Comments
"Enter comment text here"(no quotes!)

This is what it looks like:

Acme Antibodies

 * Url: Acme Antibodies

Comments
Fine supplier. Ship in a timely manner and always has been reliable.

Beta Antibodies

 * Url: Beta Antibodies

Comments
Fine supplier. Ship in a timely manner and always have been reliable.

Boiling down the template to the essentials

 * Jakob:Name and Target contain the same info. Let's drop Target and just have the name field that becomes the infobox header.
 * Target UniProt ID should be kept. It's important because the UniProt enforces a unique name and a link to more useful info.
 * Type - Not sure what this is for? What is antibody-cat1? Can we remove this or hide it?
 * Species and Description are currently also overlapping. Let's drop species since it's also confusing to have it close to the fields non/cross-reactive species.
 * Jakob: Ok, just implemented that myself. Hopefully without making a mess of the template code.

Putting parameters in order
Jakob: Suggested re-ordering of parameters:
 * 1) Antibody type - here: mouse monoclonal
 * 2) TargetID - most important info; should come top
 * 3) Raised against - should be directly after target protein because it explains whether the antibody recognises a specific peptide (part of the protein) or the full-length
 * 4) Cross-reactive (with)
 * 5) Not cross-reactive (with)
 * Bill Flanagan (OpenWetWare.org at MIT) 09:55, 22 June 2009 (EDT)Done. Check and revise full list for me as needed.
 * Bill Flanagan (OpenWetWare.org at MIT) 15:13, 22 June 2009 (EDT) Other re-ordering has been done. Please comment when you have time.
 * Jakob: Note I moved antibody type up.
 * Jakob: Ok, just implemented that myself.

Regarding experiment paraffin is generally the most challenging technique for an antibody, so it should come last after cryo.
 * Bill Flanagan (OpenWetWare.org at MIT) 09:55, 22 June 2009 (EDT)Will change
 * Jakob: Ok, just implemented that myself.

Subdividing infobox parameters
Jakob: Logically, the bottom 5 are separate from the top half - experiments vs. antibody properties. It would be great if this were highlighted graphically, e.g. by drawing an additional box around the top and bottom halves. The bottom half could have the heading: "Suitability for.." or sth similar.
 * Bill Flanagan (OpenWetWare.org at MIT) 09:55, 22 June 2009 (EDT)still working on this. box will appear soon. Templates are not easy with nested tables but not impossible
 * Bill Flanagan (OpenWetWare.org at MIT) 15:13, 22 June 2009 (EDT)Updated. I have added a header to subdivide the table rather than split to 2 tables. Looks a lot better now.
 * Jakob: Eccelente - just the look I was hoping for.

Antibody type

 * Jakob: we can either use description to contain source organisms and antibody type (e.g. mouse monoclonal IgG); antibody type would be more descriptive than description
 * OR


 * split it into 2 fields:
 * clonality: monoclonal or polyclonal (currently only antibodies raised in mice can be monoclonal, although that might change as technology advances)
 * raised in species: mouse, rat, guinea pig, rabbit, goat, chicken,..


 * Bill Flanagan (OpenWetWare.org at MIT) Need input: you tell me. Will do as you want.
 * Jakob: let's go for one field since currently all non-mouse Ab are polyclonal
 * Bill Flanagan (OpenWetWare.org at MIT) 09:55, 22 June 2009 (EDT)Done
 * Bill Flanagan (OpenWetWare.org at MIT) 15:21, 22 June 2009 (EDT)Do you still want to add clonaity (monoclonal or polyclonal)?
 * we can set the valid names to a pair but add others over time.
 * adding more values may require editing the template. will see if there's a way to use a list.
 * Bill Flanagan (OpenWetWare.org at MIT) 15:21, 22 June 2009 (EDT)Do you still want to add "raised in"?

Greek letters
Jakob The Greek letter looks strange in this font.

Bill Flanagan (OpenWetWare.org at MIT) (ΑαΒβΓγΔδΕεΖζΗηΘθΙιΚκΛλΜμΝνΞξΟοΠπΡρΣσςΤτΥυΦφΧχΨψΩω ) (&Alpha;&Beta;&Gamma;&Delta;&Epsilon;&Zeta;&Eta;&Theta;&Iota;&Kappa;&Lambda;&Mu;&Nu;&Xi;&Omicron;&Pi;&Rho;&Sigma;&Tau;&Upsilon;&Phi;&Chi;&Psi;&Omega;&alpha;&beta;&gamma;&delta;&epsilon;&zeta;&eta;&theta;&iota;&kappa;&lambda;&mu;&nu;&xi;&omicron;&pi;&rho;&sigmaf;&sigma;&tau;&upsilon;&phi;&chi;&psi;&omega;&thetasym;&upsih;&piv;)


 * Bill Flanagan (OpenWetWare.org at MIT) 09:55, 22 June 2009 (EDT)We need a better one then. Both should look at alternatives
 * Bill Flanagan (OpenWetWare.org at MIT) 17:23, 22 June 2009 (EDT)Use the buttons at the bottom of the editor bar to use the characters here rather than the html equivalents starting with an ampersand and ending with a semicolon
 * No difference. (both types shown above)
 * Jakob Minor issue; we can just leave it the way it is for the moment

Header
Jakob - I would favour anti Tubulin instead of the current Name: Tubulin. It's a frequent short hand for antibody against tubulin.
 * Bill Flanagan (OpenWetWare.org at MIT) 09:55, 22 June 2009 (EDT)Done

Parameters

 * Jakob - target: e.g. standard protein name, possibly link to UniProt or NCBI protein
 * OK
 * Jakob: put in an example target protein; abbreviation in target field, full name in box header
 * Jakob - raised against: full-length protein, peptide
 * OK
 * Bill Flanagan (OpenWetWare.org at MIT) 09:55, 22 June 2009 (EDT)Done


 * Jakob - cross-reactive against: mouse, rat, human,.. (this will always be the species it was raised against, here human and may be more species)
 * OK
 * Bill Flanagan (OpenWetWare.org at MIT) 09:55, 22 June 2009 (EDT)Done


 * Jakob - not cross-reactive against: as above
 * OK
 * Jakob: made this italic to offset it from the species where this antibody can be used
 * Bill Flanagan (OpenWetWare.org at MIT) 09:55, 22 June 2009 (EDT)Done

Usability for experiments
Types of experiments:
 * Protein blot (Western)
 * Immunoprecipitation (IP)
 * Staining fixed cells (ICC)
 * Staining paraffin sections (IHC-paraffin)
 * Staining cryo-sections (IHC-cryo)
 * Jakob: (I replaced the more cryptic names immunohistochemistry.. with more descriptive and more easily understood names)
 * Bill Flanagan (OpenWetWare.org at MIT) 09:55, 22 June 2009 (EDT)Done

Usability evaluation:
 * worked
 * uncertain
 * failed
 * untested
 * OK Coloring is automatic if one of the 4 tags above used. Otherwise it will be black. - Great, I like that.
 * Bill Flanagan (OpenWetWare.org at MIT) 09:55, 22 June 2009 (EDT)Done