Health researchers gain open access to a pan-cancer proteomics resource measuring approximately 1,500 proteins across more than 1,800 patient plasma samples, providing a detailed interactive foundation for biomarker discovery and multi-protein signature analysis.
Improving early, accurate cancer diagnosis remains a top priority for oncology clinicians and researchers. Circulating proteins occupy a critical role in these initiatives, functioning as minimally invasive biomarkers that reflect real-time biology during active disease. These protein biomarkers can reveal disease changes well before clinical diagnosis occurs, offering an avenue to improve detection accuracy and differentiate various cancer types.
To support this work, landmark proteomic datasets are now freely available to the scientific community through interactive data stories designed to facilitate the interpretation of complicated proteome datasets within current biological knowledge.
U-CAN Biobank Data Powers the Uppsala Pan-Cancer Atlas
The Disease Atlas: Cancer interactive data story stems from work led by Professor Mathias Uhlén and SciLifeLab located in Uppsala, Sweden. Their investigation executed a pan-cancer analysis examining protein associations across 15 different cancer types. The team measured approximately 1,500 proteins using plasma samples drawn from over 1,800 patients from the Uppsala-Umeå Comprehensive Cancer Consortium (U-CAN) biobank.
This platform allows investigators to examine how protein expression changes across different malignancies. Researchers can evaluate observed protein expression profiles for each cancer type, helping them pinpoint strong indicators for specific conditions.
Interactive Analytical Features in the Olink Platform
The resource provides several computational tools for users examining the proteomic findings. Investigators can select or search for proteins of interest to generate NPX distributions and receiver operating characteristic (ROC) curves.
- Apply Lasso logistic regression to identify the top 10 proteins associated with each malignancy, identifying whether they are differentially expressed or up-regulated.
- Inspect volcano plots, categorization algorithms, and Reactome pathways to gain a deeper understanding of specific cancer types.
- Analyze receiver operating characteristic (ROC) curves to determine the effectiveness of selected markers in distinguishing between cancer types.
The platform operates alongside next-generation proteomics solutions that give scientists the ability to construct their own research questions and compare their significant discoveries against premium datasets. These tools are designated for research use only and are not for use in diagnostic procedures.