The Antibodypedia  is a web service intended to provide better data on the usability of anti-protein antibodies. It's currently limited to human proteins but expansion to other species is planned (see comments). The website provides standardised results for about 963.000 antibodies against 18.900 genes covering 91% of the human genome . The antibodies are scored using an algorithm that takes into account the amount of supportive data present for the antibody . Submissions by researchers are encouraged and have to pass review. Pages are not editable but scientists can comment on individual antibodies.
Creation & Funding
Initiated by Mathias Uhlén, Erik Björling, and colleagues  from the Royal Institute of Technology in Stockholm in 2008 . The database was kick-started with about 4000 polyclonal antibodies generated by the Human Protein Atlas but now contains submissions from numerous commercial antibody vendors. The database software was improved in subsequent versions according to researcher and provider comments: versions 2 & 3 in 2009, versions 4 & 5 in 2010 , version 6 (as part of a collaboration with Nature Publishing Group) in 2011, version 7 in 2012, and version 8 in 2013. The project is funded by the 6th and 7th European Framework Programme and the Knut and Alice Wallenberg Foundation, Sweden.
Part of the results returned by Antibodypedia when queried for insulin.
P stands for experimental data from the provider (often the company selling the antibody).
U stands for user submitted data and is colour coded: green = supportive, yellow = uncertain, red = not supportive .
I asked whether the AntibodypediA will be extended to common model organisms like mouse, etc. This is the reply I received via email: "We are also planning to include mouse antibodies in Antibodypedia. We will begin working towards that during 2012. Already today we have antibodies reactive to both human and mouse. To find all antibodies stated to be reactive in mouse you can use the search ." --- JS 10:36, 20 October 2011 (EDT)
I'm still wondering whether an open submission system may result in more user contributions.