The abstract is read at the publisher; the summary is JClub's.
A 23-protein panel selected from Olink and SomaScan plasma proteomics predicts 10-year myocardial infarction risk with a 3.3-fold incidence gap between top and bottom 20% groups.
- Why it matters: Conventional risk factors and polygenic scores leave room to improve discrimination and calibration for myocardial infarction risk prediction, motivating scalable protein-based biomarkers.
- What they did: Plasma levels of >9500 proteins were measured in 4000 Chinese adults using Olink Explore and SomaScan, then machine-learning models were built and Boruta-selected 23 proteins were scored and externally tested in 37,187 Europeans from UK Biobank.
- The result: Adding the 23-protein score to conventional risk factors increased C-statistics from 0.846 to 0.864 (NRI 72%) in Chinese adults and from 0.710 to 0.733 (NRI 45%) in UK Biobank.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.