Student or educator? Sign up with your school email and get 6 months of Pro free.

The DOI matches its Crossref record; 67 of the 67 references it cites were found in a registry or an index.

Synergistic insights into human health from aptamer- and antibody-based proteomic profiling

Open access

Abstract

Affinity-based proteomics has enabled scalable quantification of thousands of protein targets in blood enhancing biomarker discovery, understanding of disease mechanisms, and genetic evaluation of drug targets in humans through protein quantitative trait loci (pQTLs). Here, we integrate two partly complementary techniques—the aptamer-based SomaScan ® v4 assay and the antibody-based Olink assays—to systematically assess phenotypic consequences of hundreds of pQTLs discovered for 871 protein targets across both platforms. We create a genetically anchored cross-platform proteome-phenome network comprising 547 protein–phenotype connections, 36.3% of which were only seen with one of the two platforms suggesting that both techniques capture distinct aspects of protein biology. We further highlight discordance of genetically predicted effect directions between assays, such as for PILRA and Alzheimer’s disease. Our results showcase the synergistic nature of these technologies to better understand and identify disease mechanisms and provide a benchmark for future cross-platform discoveries.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1038/s41467-021-27164-0

Cites

67 of 67 references verified0 citation errors · 0 not checkable

Statistics

Figures

Fig. 1 Scheme of the study design. The Venn diagram displays the overlap in protein targets captured by the SomaScan assay and the Olink proximity extension assay (PEA). Modes of binding to the protein target are depicted simplified next to each ellipse. Correlation coefficients were used to compare both technologies and factors possibly accounting for measurement differences and low correlation coefficients examined in a subset of 485 individuals with overlapping measurements. For the set of 871 common protein targets, genome-wide association analysis was performed in 10,708 (SomaScan assay) and 485 (Olink PEA) participants in the Fenland cohort. Correspondence of genetic associations was analysed by examining local genetic architecture, comparison of effect estimates and evaluation of phenotypic consequences. Parts of this figure have been created with BioRender.com.
Fig. 1 Scheme of the study design. The Venn diagram displays the overlap in protein targets captured by the SomaScan assay and the Olink proximity extension assay (PEA). Modes of binding to the protein target are depicted simplified next to each ellipse. Correlation coefficients were used to compare both technologies and factors possibly accounting for measurement differences and low correlation coefficients examined in a subset of 485 individuals with overlapping measurements. For the set of 871 common protein targets, genome-wide association analysis was performed in 10,708 (SomaScan assay) and 485 (Olink PEA) participants in the Fenland cohort. Correspondence of genetic associations was analysed by examining local genetic architecture, comparison of effect estimates and evaluation of phenotypic consequences. Parts of this figure have been created with BioRender.com.CC BY · PubMed Central
Fig. 2 Summary of correlations between measurements on both platforms. a Distribution of correlation coefficients across 937 mapping aptamer–Olink measure pairs (n = 871 unique protein targets). Source data can be found in Supplementary Data 2. b Importance measures derived from a random-forest-based variable selection procedure to predict Spearman correlation coefficients across all 871 protein targets, including technical factors and protein characteristics. Coloured boxplots indicate variables for which the importance measure remained significant after accounting for multiple testing (adjusted p < 0.01). Boxplots display the distribution of importance measures for each variable across 500 bootstrap samples. For the purpose of visualisation, median values have been highlighted by bars. % below LOD = fraction of measurement values below the detection limit of the assay. Source data are provided as a Source Data file.
Fig. 2 Summary of correlations between measurements on both platforms. a Distribution of correlation coefficients across 937 mapping aptamer–Olink measure pairs (n = 871 unique protein targets). Source data can be found in Supplementary Data 2. b Importance measures derived from a random-forest-based variable selection procedure to predict Spearman correlation coefficients across all 871 protein targets, including technical factors and protein characteristics. Coloured boxplots indicate variables for which the importance measure remained significant after accounting for multiple testing (adjusted p < 0.01). Boxplots display the distribution of importance measures for each variable across 500 bootstrap samples. For the purpose of visualisation, median values have been highlighted by bars. % below LOD = fraction of measurement values below the detection limit of the assay. Source data are provided as a Source Data file.CC BY · PubMed Central
Fig. 3 Consistency of genetic effects across platforms. a Comparison of beta estimates from linear regression models across 816 corresponding SOMAmer–Olink pairs (n = 770 unique protein targets) with at least one genome-wide associated genetic variant for either of the two, including 1267 distinct genetic variants (R2 < 0.8). Colouring is based on the genomic location of genetic variants. Red indicates variants close to the protein-encoding gene (cis, ±500 kb) and blue otherwise. Estimates are presented in Supplementary Data 3. b Summary of platform agreement for 479 genomic region–protein target associations with sufficient power among the Fenland subsample with available Olink measures (N = 485). More information is detailed in Supplementary Data 5. c Factors associated with pQTLs that are shared across platforms compared to three sets of platform-specific controls. Odds ratios and 95% confidence intervals for factors associated with cross-platform protein quantitative trait loci (pQTL) across the SomaScan v4 and Olink assays (Supplementary Data 9). The panels are based on 540 variant–protein target pairs (306 shared, 234 platform-specific) with sufficient power for replication in the Fenland sample. PAV protein altering variant, eQTL expression quantitative trait loci, Coloc. colocalisation, GWAS genome-wide association analysis. d Spearman correlation coefficients stratified by genotype. The first bar in each column indicates the overall correlation, and the three successive bars indicate the correlation among homozygous carriers of the major allele, heterozygous carriers and homozygous carriers of the minor allele (if any). Colours indicate whether the pQTL was in cis (orange) or trans (blue). Protein target–pQTL pairs were selected based on a linear regression model (see main text). Source data are provided as a Source Data file.
Fig. 3 Consistency of genetic effects across platforms. a Comparison of beta estimates from linear regression models across 816 corresponding SOMAmer–Olink pairs (n = 770 unique protein targets) with at least one genome-wide associated genetic variant for either of the two, including 1267 distinct genetic variants (R2 < 0.8). Colouring is based on the genomic location of genetic variants. Red indicates variants close to the protein-encoding gene (cis, ±500 kb) and blue otherwise. Estimates are presented in Supplementary Data 3. b Summary of platform agreement for 479 genomic region–protein target associations with sufficient power among the Fenland subsample with available Olink measures (N = 485). More information is detailed in Supplementary Data 5. c Factors associated with pQTLs that are shared across platforms compared to three sets of platform-specific controls. Odds ratios and 95% confidence intervals for factors associated with cross-platform protein quantitative trait loci (pQTL) across the SomaScan v4 and Olink assays (Supplementary Data 9). The panels are based on 540 variant–protein target pairs (306 shared, 234 platform-specific) with sufficient power for replication in the Fenland sample. PAV protein altering variant, eQTL expression quantitative trait loci, Coloc. colocalisation, GWAS genome-wide association analysis. d Spearman correlation coefficients stratified by genotype. The first bar in each column indicates the overall correlation, and the three successive bars indicate the correlation among homozygous carriers of the major allele, heterozygous carriers and homozygous carriers of the minor allele (if any). Colours indicate whether the pQTL was in cis (orange) or trans (blue). Protein target–pQTL pairs were selected based on a linear regression model (see main text). Source data are provided as a Source Data file.CC BY · PubMed Central
Fig. 4 A genetically anchored protein–phenotype network. Each node is either a protein target (square) or a phenotype (circle) and a connection was drawn between both if the protein target shared a genetic signal (posterior probability (PP) >80%) with the phenotype within a ±500 kb region around the protein-encoding gene (Supplementary Data 11). Colours indicate whether the shared signal was identified using the SomaScan (cyan) or the Olink assay (orange). Protein–phenotype connections consistent between both platforms are indicated by two lines connecting the protein and the phenotype. Solid lines indicate a positive association of the shared genetic variant with the phenotype aligned to the protein-increasing allele. The inset highlights selected subnetworks for which both proteomic techniques provide complementary information.
Fig. 4 A genetically anchored protein–phenotype network. Each node is either a protein target (square) or a phenotype (circle) and a connection was drawn between both if the protein target shared a genetic signal (posterior probability (PP) >80%) with the phenotype within a ±500 kb region around the protein-encoding gene (Supplementary Data 11). Colours indicate whether the shared signal was identified using the SomaScan (cyan) or the Olink assay (orange). Protein–phenotype connections consistent between both platforms are indicated by two lines connecting the protein and the phenotype. Solid lines indicate a positive association of the shared genetic variant with the phenotype aligned to the protein-increasing allele. The inset highlights selected subnetworks for which both proteomic techniques provide complementary information.CC BY · PubMed Central
Fig. 5 Regional association plots for the PILRA and GDF15 locus. a Regional association plots for paired immunoglobulin-like type 2 receptor alpha (PILRα) measured by SomaScan (top rows) and Olink, as well as for Alzheimer’s disease centred around a colocalising signal for the missense variant rs1859788 within PILRA (p.G78R). Colours indicate direction of effect for the A-allele of rs1859788 on the respective trait (blue—inverse, red—positive) and shading indicates linkage disequilibrium (r2) with the lead variant at the locus. The red line in the gene panel indicates the position of the variant. P values for protein measures were derived from genome-wide association analysis (linear regression models) from the Fenland cohort as described in the main text, whereas summary statistics for Alzheimer’s disease was obtained from Jansen et al.58. The scheme on the right displays a possible mechanism by which the p.G78R could lead to discordant results between SomaLogic and Olink. b Each panel shows summary statistics (p values) from genetic association studies for coffee intake, comparative body size at age 10, body mass index (conditioned on lead signals), and growth-differentiation factor 15 (GDF-15) measured by Olink and SomaScan. The lead genetic variants for each assay as well as variants in high linkage disequilibrium are highlighted by colours (blue—SomaScan, orange—Olink). Summary statistics for phenotypes were obtained from the Open GWAS database (UK Biobank)55 and protein summary statistics for GDF-15 from8 for Olink and SomaScan from the present study. The scheme on the right shows possible consequences of a differently shaped GDF-15 protein. A 3D model of the alternative protein is presented in Supplementary Fig. 10. Parts of this figure have been created with BioRender.com. Source data are provided as a Source Data file.
Fig. 5 Regional association plots for the PILRA and GDF15 locus. a Regional association plots for paired immunoglobulin-like type 2 receptor alpha (PILRα) measured by SomaScan (top rows) and Olink, as well as for Alzheimer’s disease centred around a colocalising signal for the missense variant rs1859788 within PILRA (p.G78R). Colours indicate direction of effect for the A-allele of rs1859788 on the respective trait (blue—inverse, red—positive) and shading indicates linkage disequilibrium (r2) with the lead variant at the locus. The red line in the gene panel indicates the position of the variant. P values for protein measures were derived from genome-wide association analysis (linear regression models) from the Fenland cohort as described in the main text, whereas summary statistics for Alzheimer’s disease was obtained from Jansen et al.58. The scheme on the right displays a possible mechanism by which the p.G78R could lead to discordant results between SomaLogic and Olink. b Each panel shows summary statistics (p values) from genetic association studies for coffee intake, comparative body size at age 10, body mass index (conditioned on lead signals), and growth-differentiation factor 15 (GDF-15) measured by Olink and SomaScan. The lead genetic variants for each assay as well as variants in high linkage disequilibrium are highlighted by colours (blue—SomaScan, orange—Olink). Summary statistics for phenotypes were obtained from the Open GWAS database (UK Biobank)55 and protein summary statistics for GDF-15 from8 for Olink and SomaScan from the present study. The scheme on the right shows possible consequences of a differently shaped GDF-15 protein. A 3D model of the alternative protein is presented in Supplementary Fig. 10. Parts of this figure have been created with BioRender.com. Source data are provided as a Source Data file.CC BY · PubMed Central
Fig. 6 Sources of variation. Graphical summary of factors contributing to variation in the affinity-based discovery of the plasma proteome. PAV protein altering variant, SNV single-nucleotide variant. This figure has been created with BioRender.com.
Fig. 6 Sources of variation. Graphical summary of factors contributing to variation in the affinity-based discovery of the plasma proteome. PAV protein altering variant, SNV single-nucleotide variant. This figure has been created with BioRender.com.CC BY · PubMed Central

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

5 more claims about this paper

Create a free account to see every claim other papers make about this one.

See all claims — free account

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 67 references

Cited by

See all 236

Metadata

Authors
Pietzner, M., Wheeler, E., Carrasco-Zanini, J., Kerrison, N. D., Oerton, E., Koprulu, M., Luan, J., Hingorani, A. D., Williams, S. A., Wareham, N. J., & Langenberg, C.
Affiliations
  • University of Cambridge (GB)
  • Health Data Research UK (GB)
Corresponding author
Langenberg, Claudia
Venue
Nature Communications
Published
2021-11-24
Volume
12
Issue
1
Pages
6822-6822
ISSN
2041-1723 (ISSN-L)
OpenAlex
W3212460220
Cited by
234
References
67
Citations by year
2026: 55 · 2025: 49 · 2024: 55 · 2023: 44 · 2022: 29 · 2021: 2
FWCI
18.50
Type
journal article
Language
en
Open access
Licence
Locations
  • Nature Communications (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • UCL Discovery (University College London) (submittedVersion)
  • Queen Mary Research Online (Queen Mary University of London) (submittedVersion, cc-by)
  • Apollo (University of Cambridge) (submittedVersion)
  • Apollo (University of Cambridge) (submittedVersion, cc-by)
  • Apollo
  • Apollo (cc-by)
Funders
  • Wellcome Trust
  • UK Research and Innovation (MR/V033867/1)
  • National Institute for Health and Care Research (MC_UU_12015/1, NIHR202383)
  • British Heart Foundation (AA/18/6/34223)
  • University of Cambridge (00001)
  • Cambridge Trust
  • Medical Research Council (MC_UU_12015/1, MC_PC_13046, MC_UU_12015, MR/V033867/1)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Genetic Associations and Epidemiology
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Advanced Biosensing Techniques and Applications
  • Physical Sciences › Chemistry › Spectroscopy › Advanced Proteomics Techniques and Applications
Keywords
protein quantitative trait loci, pQTL, aptamer-based proteomics, SomaScan, Alzheimer's disease, biomarker discovery, disease mechanisms
MeSH terms
Adult, Alzheimer Disease, Antibodies, Female, Humans, Male, Membrane Glycoproteins, Middle Aged, Phenotype, Receptors, Immunologic, Cohort Studies, Proteome, Protein Interaction Mapping, Quantitative Trait Loci, Proteomics, Aptamers, Peptide, Protein Interaction Maps

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Synergistic insights into human health from aptamer- and antibody-based proteomic profiling. https://app.cochat.ai/papers/10.1038/s41467-021-27164-0
The paper
Pietzner, M., Wheeler, E., Carrasco-Zanini, J., Kerrison, N. D., Oerton, E., Koprulu, M., Luan, J., Hingorani, A. D., Williams, S. A., Wareham, N. J., & Langenberg, C. (2021). Synergistic insights into human health from aptamer- and antibody-based proteomic profiling. Nature Communications. https://doi.org/10.1038/s41467-021-27164-0

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.