Cross-platform proteomics to advance genetic prioritisation strategies

Abstract

Discovery of protein quantitative trait loci (pQTLs) has been enabled by affinity-based proteomic techniques and is increasingly used to guide genetically informed drug target evaluation. Large-scale proteomic data are now being created, but systematic, bidirectional assessment of platform differences is lacking, restricting clinical translation. We compared genetic, technical, and phenotypic determinants of 871 protein targets measured using both aptamer-(SomaScan® Platform v4) and antibody-based (Olink) assays in up to 10,708 individuals. Correlations coefficients for overlapping protein targets varied widely (median 0.38, IQR: 0.08-0.64). We found that 64% of pQTLs were shared across both platforms among all identified 608 cis- and 1,315 trans-pQTLs with sufficient power for replication, but with correlations of effect estimates being lower than previously reported (cis: 0.41, trans: 0.34). We identified technical, protein, and variant characteristics that contributed significantly to platform differences and found contradicting phenotypic associations attributable to those. We demonstrate how integrating phenomic and gene expression data improves genetic prioritisation strategies, including platform-specific pQTLs.

A genome-wide association study of serum proteins reveals shared loci with common diseases

“…The majority of cis and trans acting pQTLs detected in serum and plasma can be readily replicated across different populations, as shown in the current study, and different proteomic platforms 8,9,17,21 . However, a recent cross-platform comparison has shown that a subset of pQTLs are platform-specific and may in some cases represent epitope effects or other technical factors 21 .…”

“…The majority of cis and trans acting pQTLs detected in serum and plasma can be readily replicated across different populations, as shown in the current study, and different proteomic platforms 8,9,17,21 . However, a recent cross-platform comparison has shown that a subset of pQTLs are platform-specific and may in some cases represent epitope effects or other technical factors 21 . Thus, meta-analyses across platforms will still need to consider differences in analytical approaches and in cases where protein quantifications obtained by orthogonal methods differ, cis-pQTLs and mass spectrometry validation of probe targets may be good indicators of platform specificity 34 .…”

“…S5-S6). This locus has in fact been identified as platform specific in a recent study 21 and was suggested to be related to white blood cell lysis during sample handling. We similarly performed a lookup of the independent study-wide significant associations identified in the current study in the INTERVAL study summary statistics (Supplementary Note 2, Fig.…”

“…Previous studies have shown that pQTLs replicate well across proteomics platforms 8,9 . While a recent comparisons of protein measurements across different platforms showed a wide range of correlations 21,34 , cis pQTLs and validation by mass spectrometry were predictive of a strong correlation across platforms and are likely good indicators of platform specificity when protein concentrations obtained by orthogonal methods differ 34 . Hybridization controls were used to correct for systematic variability in detection and calibrator samples of three dilution sets (40%, 1% and 0.005%) were included so that the degree of fluorescence was a quantitative reflection of protein concentration.…”

Towards Building a Quantitative Proteomics Toolbox in Precision Medicine: A Mini-Review

“…However, they have several limitations which include as: no detection of proteins that are not targeted by the assay (comprehensiveness); binding affinity differences across proteins or non-specific binding for variant proteins (quantitative accuracy); and no distinction between posttranslational modified proteins and isoforms (specificity; Yeh et al, 2017;Raffield et al, 2020). For SOMAScan and Olink, Pietzner et al (2021) showed that factors of technical variability can be introduced by target proteins with transmembrane domain, glycosylation effects, or protein-altering variants (Pietzner et al, 2021). Still, implementations of multiplexed platforms into clinical settings are relatively new, given that more research and verification are still needed to validate these as clinical-grade technologies (Williams et al, 2019).…”