The Human Protein Atlas

The Human Protein Atlas

The Human Protein Atlas project was initiated in 2003 by Swedish researchers, headed by Professor Mathias Uhlén, and funded by the Knut and Alice Wallenberg foundation. It is a unique world leading effort performing systematic exploration of the human proteome using antibodies.

The aim of the Human Protein Atlas project is to present an expression map of the complete human proteome in 2015, with a curated version in 2020. To accomplish this, highly specific premium antibodies are developed to all protein coding human genes and protein profiling is established in a multitude of tissues and cells using tissue arrays. Expression data is available from application-specific validation. Applications applied are immunohistochemistry, Western blot analysis, protein array assay and immunofluorescent based confocal microscopy.

The antibodies developed within the Human Protein Atlas project are carefully designed and manufactured to achieve the very highest level of specificity, reliability and versatility. They are available from Atlas Antibodies, and we have named them Triple A Polyclonals. Read more about the very special story of Atlas Antibodies and the Human Protein Atlas here.

While the antibodies are available for purchase via Atlas Antibodies, all protein expression profiling data is freely available on the Human Protein Atlas portal. The valuable information within the database can serve many purposes, such as aiding in the basic understanding of different cancers, facilitating development of new tools for diagnostics and treatments can be developed and introduced into clinical medicine.

The New Human Protein Atlas, December 2013

The 12th version of the Human Protein Atlas includes antibodies targeting proteins from more than 16,600 human genes, corresponding to 82% of the human proteome. With this update, the Human Protein Atlas is now close to the goal of providing profiling data for all human proteins.

In the new version, the extensive amount of data has been divided into four separate 'sub atlases': the Tissue Atlas, the Cancer Atlas, the Subcell Atlas and the Cell Line Atlas. The four sections have different aims of showing gene expression data and making the data easy to access, search and navigate. 

Human Protein Atlas

For all proteins represented in the Tissue Atlas the expression profiles are based on immunohistochemistry (IHC) from a large number of human tissues. A very important addition in this new version is the presentation of protein expression data in correlation to RNA sequencing data for each gene. In the Cancer Atlas, differentially expressed genes in several cancers can be studied, while the Subcell Atlas presents subcellular localization by confocal microscopy. Additional information about protein expression in common cell lines is included in the Cell Line Atlas, which has become an appreciated toolbox for research.

An Atlas of Human Protein Expression

The human genome project provided the basic blueprint for the approximately 20,000 human genes present in all cell types that constitute human organs and tissues. However, what really define the function of a given cell are the amounts, localization, interactions and modifications of its expressed proteins. Immunohistochemistry (IHC) is an invaluable tool for in situ visualization of protein expression patterns in tissues and cells, and potentially presents us with an additional layer of information regarding normal cellular functions, and pathogenic mechanisms leading to different types of disease.

Normal Tissue IHC based Expression Profiles

Within the Human Protein Atlas (HPA) project, antibodies against human gene products are used to stain tissue micro arrays (TMAs) containing samples from 48 different normal human tissues, 20 different cancer types and 44 different human cell lines. The 48 normal tissues are present in triplicate samples and represent 82 different cell types. All normal tissue images have undergone pathology-based annotation of expression levels and are displayed on the normal tissue atlas presenting information regarding the expression profiles of human genes both on mRNA and protein level. The mRNA expression data is derived from deep sequencing of RNA (RNA-Seq) from 27 major different normal tissue types and the protein data covers 16.621 genes (82%) for which there are available antibodies.

The Cancer Atlas contains gene expression data based on protein expression patterns in a multitude of human cancer specimens. Altogether 216 different cancer samples, corresponding to the 20 most common forms of human cancer, have been analyzed for all included genes. All cancer tissue images have been manually annotated by pathologists and just as for the normal tissue atlas, protein data includes protein expression levels corresponding to 16.621 genes for which there are available antibodies.

In cases where data is available from two or more antibodies directed against the same target, the expression profiles are manually curated to produce an annotated protein expression profile. By taking into consideration also other sets of independently generated data the most reliable estimation of the true protein expression is obtained. The long-term goal is to provide such analyses based on two or more independent antibodies for the complete human proteome.

Tissue Atlas normal tissue stainings for ADH1A
Cancer Atlas cancer tissue stainings for ADH1A Part of the normal tissue summary view from the Tissue Atlas is shown on top. In this view, RNA expression is displayed together with protein expression for normal tissues. The image below shows the cancer tissue summary view from the Cancer Atlas for the same protein, where the intensity of the stainings of a specific antibody in cancer tissues are displayed. 


Subcellular Expression Profiles

In addition to IHC-based profiling, many of the antibodies are also used for immunofluorescence-based confocal microscopy in three cell lines of different origins. The results are shown in the subcellular protein atlas that displays high resolution, multicolor images of immunofluorescently stained cells. This analysis increases the spatial resolution of the protein localization data, and will be used to create a map of the fine cellular and subcellular expression profiles of all human proteins. For each antibody two suitable cell lines from a human cell line panel are selected for the immunofluorescence analysis, based on RNA sequencing data. The third human cell line chosen for each antibody is always U-2 OS, in order to localize the whole human proteome on a subcellular level in one cell line.

Cell Line Expression Data

The Cell Line Atlas contains information about gene expression both on mRNA and protein level in a diverse panel of cell lines. Protein data is also available for 10 patient blood samples representing normal PBMC and different types of leukemia. Protein expression is available as quantified immunohistochemical positivity in cells and cell lines assembled in microarray format. Altogether there is transcript data for 44 cell lines, while protein data is available for 21% of all protein coding genes in 46 cell lines and generated using 5272 antibodies.

A Tool For Proteome Research

The unique approach taken by the Human Protein Atlas project will allow us for the first time to address certain fundamental biological questions such as: How large is the fraction of our genome that encodes for cell type specific proteins in relation to house keeping proteins, or proteins differentially expressed across different cell types? How does this correlate with defined cell functions, and molecular and morphological pathways? With the Human Protein Atlas we can now further investigate and understand human biology.

Atlas Antibodies is working closely together with the Human Protein Atlas, and the valuable information it contains is used as an important source to find biomarkers. By exploring the information in the Human Protein Atlas, differentially expressed proteins can be identified.  These proteins are potentially interesting biomarkers that may be used for accurate diagnosis, prognosis, or selection of individual treatments. The DNA-binding protein RBM3 is an example of an oncology biomarker that has been identified via the Human Protein Atlas. The levels of RBM3 expression has been found to have a significant correlation to patient survival in several cancers. Read more about this under Oncology Biomarkers.

Explore tissue histology with the Tissue Dictionary

The Tissue Dictionary is another useful tool in the Human Protein Atlas. It contains high-resolution images of both full tissue sections stained with hematoxylin-eosin and confocal microscopy images using different probes to highlight the organelle structure of a cell. The Tissue Dictionary is aiming not only to facilitate the interpretation and use of data available in the Human Protein Atlas, but also to serve as a tool for basic training and understanding of human tissue histology, pathology and cell biology.

Forgot your email?