Normalization to external reference for reduction of technical variation
Background
Mitigation of technical variability is a key aspect in relating proteomic biomarkers to clinical endpoints. A primary source of variability comes from inherent issues in the actual technology used to measure the protein levels in a sample, but additional variability can come from many different sources, including plate differences, sample handling differences, and study or cohort differences. Without bridging samples, these different sources of variation can lead to difficulties in combining different cohorts and across studies longitudinally, preventing researchers from sharing data easily. Here, we present results for normalization to an external reference population, which allows us to combine datasets without the need for bridging samples.
Methods
SomaLogic performs a series of steps to standardize the SomaScan® Assay data. These steps are hybridization normalization, plate scaling and calibration, and the adaptive normalization by maximum likelihood (ANML), which normalizes SomaScan EDTA plasma measurements to a healthy U.S. population reference. Because this normalization technique does not require bridging samples, it can be used to combine data from different times and sources. The impact of this type of normalization is quantified through CV calculations over multiple runs and time points. As an example, we demonstrate the benefits of this method on age and sex as clinical endpoints.
Results
The median coefficient of variation (CV) on raw SomaScan data is 22.4%, which drops to 5.3% after all standardization steps and ANML normalization are applied. Longitudinally, the median CV after ANML normalization remained steady at 5.5% across all cohorts that were run between June of 2020 and May of 2023. Examination of analytes related to age and sex confirmed that known biomarkers were more easily identified after ANML normalization was applied.
Conclusions
Use of a reference population allows for bridging across varied cohorts and studies, which allows for increased use of SomaScan data
among researchers using proteomics for biomarker discovery and clinical development.
Authors
Y. Hagar
D. Maxwell
C. Jamison
M. Westacott
D. Perry
SomaLogic Operating Co., Inc., Boulder, CO, USA
Learn more by downloading this poster
More posters
PosterComparison of Proteomic CV Risk to Established ASCVD 10-Year Risk Decision Points
The ASCVD pooled cohort equation (PCE) is well-established for CV risk assessment. Decision points for determining treatment plans are low, intermediate and high risk over 10 years, however this approach over and underestimates risk in certain subgroups. The validated CV Risk SomaSignal® Test (SST) provides 4-year risk probability of MACE allowing for timely assessment of risk, but the shorter timescale makes comparison to 10-year PCE risk less intuitive.
PosterStatin signature: using proteomics to detect pharmacological fingerprints
Using a previously described metacohort (n=5,575) of patients with increased CV risk, we hypothesized that PCE would stratify patients differently than the CV Risk SST, and that CV Risk score scaled to 10 years would yield an improved net reclassification index (NRI).
PosterUsing a proteomics-based cardiovascular risk test to identify systemic changes in a clinical trial of nonalcoholic fatty liver disease
Improvement in hepaKc inflammaKon, NAFLD acKvity score and fibrosis were associated with improved proteomic CV risk scores regardless of treatment provided.