Biowire Header


Biowire Winter 2011 — Knockout Rats — More Effective Pre-Clinical Models

Table of Contents


The New Human Protein Atlas Web Portal

  • Coverage of >10,000 protein coding genes
  • Gene-centric data presentation
  • Knowledge-based annotated protein expression
  • Summarized view
  • RNA data added
  • New search functions
  • New logotype and design of the Human Protein Atlas

Protein Expression Profiles for 10,000 Genes 2010

The first version of the Human Protein Atlas was released in 20051. This first version included protein expression data from 718 antibodies corresponding to 650 human proteincoding genes. High-resolution immunohistochemistry images from a multitude of normal and diseased human tissue samples were presented.

In November 2010, a milestone was reached within the Human Protein Atlas project ( with the representation of protein expression data for more than 10,000 genes corresponding to 50% of the human genome2 (Table 1). For example, extensive protein profiling data from almost 80% of the human kinases and more than 50% of the transcription factors are presented on the portal. Protein expression is validated using techniques such as immunohistochemistry (IHC), immunofluorescence (IF), and Western Blot (WB). IHC images covering the majority of human normal and cancer tissues, as well as a large selection of cell lines and primary cells, are publicly available for each gene3. Subcellular analysis using confocal microscopy is performed, and the resulting IF images are presented on the portal4.

Each year, protein expression and localization data of ~2,500 new proteins are added to the portal. By 2015, a first draft of the localization profile of the whole human proteome will be completed.

Table 1. Number of genes with protein expression profiles in version 1.0 - 7.0 of the Human Protein Atlas.

Version Year Number of Genes
1.0 2005 650
2.0 2006 1,344
3.0 2007 2,600
4.0 2008 5,067
5.0 2009 6,844
6.0 2010 8,489
7.0 2010 10,120


Annotated Protein Expression

One of the major objectives in the 7.0 release is to move towards a knowledge-based presentation of protein expression data. In this attempt, several antibodies with non-overlapping epitopes towards the same protein target are used to validate each other´s expression profiles. This validation results in an “annotated expression profile.” Microscopic images from a multitude of human tissues and cells are reviewed simultaneously, and all data is assembled using updated experimental or predicted bioinformatic data as a guide. Protein expression data is thus presented gene-centric instead of antibody-centric, resulting in a single annotated expression profile on the portal for each protein-coding gene. The annotation process also includes review of possible cross-reactivity, data from Western Blot, immunofluorescence analysis, transcript profiling, and/or siRNA knockdowns. All antibodies directed against targets from a particular gene are combined into one page. A reliability score is generated based on the degree of knowledge-based certainty. When only one antibody is available for a protein target, no annotated protein expression is generated, and the results from the corresponding gene will be reported only as “antibody staining.”

Summarized View

Each human gene has a summary view on the new portal to allow for a convenient entry point to navigate to the more detailed pages. On the summary page, the annotation results from one or several antibodies against a specific target protein are presented together with links to the pages containing the original high-resolution images. Figure 1 shows a part of the summary view for the human Antigen KI-67 (MKI67). In addition, the summary view for each gene includes Gene/Protein Summary, Antibody/ Antigen Summary, Protein Atlas Summary, and Cell Line/Cell Sample Summary.

Human Protein Atlas Figure 1
Figure 1. Part of the summary view for the human Antigen KI-67 (MKI67) on the Human Protein Atlas portal, displaying expression data for the target protein. The figure shows summary of subcellular localization and of protein expression in normal and diseased human tissues. Not shown in this figure is the summary of expression in cell lines and cell samples.

Detailed View

From the summary view, there are links to detailed pages describing protein expression in normal tissues and organs, in cancer tissues, in cell lines, and in subcellular compartments. The detailed IHC page of normal human tissues has been redesigned to an organ view to enable the cell types of similar origin to be easily compared. The tissues and organs are divided into 12 functional classes from which a total of 66 normal cell types from 46 different tissues and organs are being scored. Figure 2 shows a part of the expression profiles for Antigen KI-67 (focused on the central nervous and hematopoietic systems), including the antibody staining results from three independent anti-MKI67 antibodies and the resulting annotated protein expression. When clicking on a specific tissue, information from the IHC stainings of the selected tissue using the anti-MKI67 antibodies will appear. Shown are high-resolution IHC images from three different individuals and information of cell types, staining intensities, localizations, and tissue material.

Human Protein Atlas Figure 2
Figure 2. Part of detailed organ view of the staining pattern of Antigen KI-67. The three red color scale columns represent the antibody staining from three independent anti-MKI67 antibodies and the blue color scale column the resulting annotated protein expression. The color intensity represents the level of expression.

RNA Expression Analysis

The results from RNA expression analysis for each gene have been added in an RNA section. The same three human cell lines used for immunofluorescence analysis, U-2 OS, A-431, and U-251 MG, are being analyzed by RNA-Seq to calculate the RNA levels. RPKM values (number of reads per kilobase gene model and million reads) are used to estimate the expression of each gene in each of the cell lines. The RNA abundance is reported as high, medium, low, or not detected. The combined result from the three cell lines is used to classify the gene into five categories: similar expression, substantially different, slightly different, cell-type specific (only detected in one cell line), and not detected (not detected in any cell line).

Search Functions

The advanced search function allows complex queries involving normal and cancer tissues, subcellular localizations, and validation results for antibody and/or target protein classes. In addition, this new portal has improved search result listings with relevance-sorting and customizable displays.

Availability of Antibodies

Antibodies presented on the Human Protein Atlas are available for purchase. Information about antibody suppliers can be found on the Antibody/Antigen Summary page.

To discover more, click here.


  • Release 7.0 of the Human Protein Atlas contains protein expression profiles for more than 50% of all human proteins.
  • >10,000 Prestige Antibodies corresponding to >8,400 human proteincoding genes have characterization data on the Human Protein Atlas portal.
  • The new Human Protein Atlas portal shows gene-centric protein expression profiles based on the annotation of one or several antibodies towards the same protein target.
  • “Annotated expression” exists for target proteins analyzed by several independent antibodies. For protein targets characterized with a single antibody, the expression is denoted “antibody staining.”
  • RNA expression data from three human cell lines is presented for all genes analyzed using Prestige Antibodies.


  1. Uhlén, M. et al. A Human Protein Atlas for Normal and Cancer Tissues Based on Antibody Proteomics. Mol. Cell Proteomics 2005 4(12):1920-1932.
  2. Uhlén M, et al. Towards a knowledge-based Human Protein Atlas. Nat. Biotechnol. 2010 28(12):1248-50.
  3. Pontén F, et al. The Human Protein Atlas — a tool for pathology. J. Pathology 2008 216(4):387-93.
  4. Barbe L, et al. Toward a Confocal Subcellular Atlas of the Human Proteome. Mol. Cell Proteomics 2008 7(3):499-508.

Biowire homepage
back to top