Human Protein Atlas

What Human Protein Atlas is

The Human Protein Atlas (HPA) is a Swedish-based research programme established in 2003 to map all human proteins across cells, tissues, and organs. Directed by Professor Mathias Uhlén, the project is primarily based at the KTH Royal Institute of Technology, with collaborative research groups at Uppsala University, Karolinska Institutet, Chalmers University of Technology, and Lund University. It is funded by the Knut and Alice Wallenberg Foundation and recognised as an ELIXIR European core resource and a Global Core Biodata Resource.

The atlas addresses the challenge of understanding where and to what extent human protein-coding genes are expressed throughout the body. By combining transcriptomics, antibody-based imaging, and mass spectrometry-based proteomics, the project provides a unified reference of expression patterns, spatial localization, molecular pathways, and structural annotations across healthy and diseased human biology.

Scientists can use the atlas to explore human protein functions, identify tissue-specific biomarkers, investigate cell cycle variations, and evaluate the prognostic relevance of genes across various cancers using both experimental measurements and predicted protein structures.

How it is organised

The database integrates data from several experimental assays, including conventional and multiplex immunohistochemistry (IHC), immunocytochemistry and immunofluorescence (ICC-IF) using confocal microscopy, Western blotting, protein arrays, single-cell RNA sequencing (scRNAseq), single-nuclei RNA sequencing (snRNAseq), and mass spectrometry (MS), including Deep Visual Proteomics (DVP). Antibody reagents are validated according to criteria suggested by the International Working Group for Antibody Validation (IWGAV).

Data is curated and compartmentalised into nine main resources: Tissue, Brain, Single Cell, Cancer, Blood, Subcellular, Cell Line, Structure, and Interaction. In addition to antibody-derived imaging and curated transcriptomic data, the platform integrates computational models, including three-dimensional protein structures generated using the AlphaFold source code, mutation effect predictions from AlphaMissense, and metabolic networks from the Metabolic Atlas.

Key facts

What you can use it for

Evaluating tissue-specific protein expression

Researchers can inspect protein and RNA expression across 45 normal human tissues using annotated immunohistochemistry slide images and transcriptomic specificity categories.

Investigating subcellular protein localisation

Cell biologists can determine organelle distribution across up to 49 subcellular compartments and assess cell cycle-dependent variations in protein expression.

Assessing cancer prognostic markers

Oncologists can examine the correlation between mRNA expression levels of individual human genes and patient survival through Kaplan-Meier survival plots across 21 cancer types.

Cross-species mammalian brain comparisons

Neuroscientists can study spatial protein and RNA distributions across 13 major brain regions and regional subfields in humans and compare them with orthologues in pigs and mice.

Analysing 3D structures and genetic variants

Structural biologists can interact with AlphaFold-predicted models to inspect protein domains, antigen sequences, transmembrane segments, and AlphaMissense pathogen predictions.

Query it here

The research agent queries Human Protein Atlas directly — no account, no API key — and every answer links back to the record it came from. Ask in plain language and it writes the query.

Strengths and limits

Questions