Plasma proteomic comparisons change as coverage expands for SomaLogic and Olink
Level 4 - case-series / case-control
Cross-sectional split-sample technical reproducibility study (graded by design analogy for assay validation)
OpenAlex W4400577998 · doi:10.1101/2024.07.11.24310161
What was done
Assessed the technical reproducibility and cross-platform agreement of the expanded plasma proteomic platforms SomaScan 11k (>11,000 aptamer assays) and Olink Explore HT (>5,400 antibody proximity extension assays). Split plasma samples from 102 Atherosclerosis Risk in Communities (ARIC) Study participants were measured to determine technical coefficients of variation (CV), relation of CV to limit of detection (LOD), and between-platform correlation across 4,443 overlapping target assays.
What was found
SomaScan 11k assays showed a median CV of 6.8% (versus 6.6% for assays on the SomaScan 5k subset). Olink Explore HT assays showed a median CV of 35.7% (versus 19.8% for assays on the Olink Explore 3072 subset). Olink CVs were strongly negatively correlated with the percentage of samples above the limit of detection. Across 4,443 overlapping assays, between-platform correlations exhibited a bimodal distribution with one peak near r ~ 0 and a smaller peak near r ~ 0.8.
Why it matters
Demonstrates that as multiplex proteomic platforms expand coverage, technical precision diverges significantly between platforms, and correlation between overlapping targets is low for many proteins.
Limits
Based on split plasma from a single cohort of 102 participants. The abstract assesses technical precision and correlation only, without evaluating analytical specificity, epitope validation, or biological validity. Published as a preprint.
Cited by
- supports Modern proteomic platforms can measure the concentrations of up to 11,000 proteins from a single drop of blood.