Proteome-Wide Production of Antibodies within the Human Protein Atlas Project

By: Jenny Ottosson Takanen, Holger Berling, Hanna Tegel, Åsa Sivertsson, Peter Nilsson, Mathias Uhlén, Poster - The Human Protein Atlas
KTH Royal Institute of Technology, School of Biotechnology Div. Proteomics
AlbaNova University Center, SE-106 91 Stockholm, Sweden

Download the poster (4.3 Mb PDF)

In the Human Protein Atlas (HPA) more than 70% of the human genes are soon portrayed using antibodies. To make this possible, more than 35000 different human protein fragments have been immunized to produce polyclonal antibodies which are then strictly scrutinized prior to their use in the protein expression and localization studies of the atlas. All genes have been initiated in the production pipeline and a strong effort on iterations, of those genes where either antigen production has failed or the antibodies did not pass the thorough HPA validation regime, is in progress. It is our experience that re-immunization of the same immunogen or selection of another fragment of a particular gene is by far the most efficient iteration strategies to gain gene coverage. So far, around 75000 fragments have been selected based on uniqueness relative to proteins from other genes. However, optimization of the protocols for cloning as well as protein expression has been essential for the successful production of immunogens for many of the more challenging genes/proteins and will continue to be so as we approach full coverage and the most demanding genes remain.

Iteration strategy in the HPA production pipeline

Iteration strategy in the HPA production pipeline.
Primary iteration shown in green: Antigens that fail to generate approved antibodies are re-immunized up to 3 times. Failed protein production lead to iteration with an optimized expression protocol. Selection of another protein fragment is the most efficientinitial iteration of genes where no expression clone is achieved with the standard protocol based on RT-PCR from RNA-pools.

Secondary iteration shown in red: Selection of another human protein fragment when optimized protein expression or reimmunization fails. Optimized cloning protocols or PCR from available cDNA for the few genes where no expression clone has been achieved with the standard protocol.

Coverage of the proteome by HPA production module

Coverage of the proteome by HPA production module.
For practically all genes (99%), at least one protein fragment has been selected and antigen production been initiated. Presently, for 89% of the proteome a polyclonal antibody with specificity verified on an antigen protein array has been aquired.

Proteome coverage

The goal to have a first draft of the human proteome portrayed with antibody proteomics based on at least one primary antibody is appraoching. For 62% of the proteome an in-house produced antibody is represented in the HPA database, and only 21% remain in antibody generation. Along with this much effort is put into generating paired antibodies to validate the results of the first.

A strict strategy has been set up in order to keep up the antibody production and yet prioritize the production of primary antibodies for the first draft. At several key points of the production the samples are strictly prioritized to give precedence to samples more likely to generate new data on a human protein not yet portrayed in the Human Protein Atlas.



  • Uhlén M. et al., Towards a knowledge-based Human Protein Atlas. Nat. Biotechnol. 2010 28(12):1248-50.
  • Uhlén M. et al., A human protein atlas for normal and cancer tissues based on antibody proteomics. Mol. Cell. Proteomics. 2005 4(12):1920-32.
  • Tegel H. et al. , High-throughput protein production – lessons from scaling up from 10 to 288 recombinant proteins per week. Biotechnol. J. 2009 4(1):51-7.


Related Links