| Title: | Integrating and Validating Human Disease Blood Atlas with UK Biobank Plasma Profiles |
| Lead Institution: | KTH Royal Institute of Technology |
| Principal investigator: |
Application 174156
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
About
To analyze this vast and complex dataset, we will employ advanced data-driven methodologies, including machine learning and artificial intelligence. These techniques are useful for handling large volumes of data, identifying patterns, and making predictions based on intricate data relationships. Machine learning algorithms will be used to sift through the protein profiles and patient information, uncovering new insights into disease markers and their interactions with various physiological factors. By training the algorithms on the UK Biobank dataset, we aim to create models that are more precise in their predictions. This is key to ensuring that our models remain relevant and accurate as new data becomes available in the future. Significance and UK Biobank Dataset Integration The UK Biobank dataset, known for its breadth and depth, has been a cornerstone in several high-impact studies published in journals like Science. Its extensive plasma profiles encompass a broad spectrum of diseases and a diverse patient demographic. By integrating this dataset with our findings from the Human Protein Atlas, we aim to validate and refine our protein signatures, enhancing the predictive accuracy for early disease detection based on clinical data (Health Outcome phenotypes). This cross-referencing is pivotal in establishing robust, universally applicable biomarkers. Preliminary Results Our pilot study, involving 1,477 patients across 12 cancer types from the U-CAN biobank, utilized an earlier version of OLINK Explore, targeting approximately 1,500 proteins. The application of machine learning techniques enabled us to develop a sophisticated classification model based on 83 upregulated proteins, significantly enhancing diagnostic accuracy. The success of this pilot study has been instrumental in shaping the methodology and objectives of our current expansive analysis.1 Publication
| Pub ID | Title | Author(s) | Year | Journal |
|---|---|---|---|---|
| 17104 | A human pan-disease blood atlas of the circulating proteome | María Bueno Álvez (+110) | 2025 | Science |
Enabling scientific discoveries that improve human health