Large-scale plasma proteomics comparisons through genetics and disease associations

Area of Science:

• Proteomics
• Genomics
• Biomedical Research

Background:

• High-throughput proteomics, genomics, and phenotypic data integration can elucidate genome-disease links.
• The UK Biobank Pharma Proteomics Project provides extensive plasma proteomics data.

Purpose of the Study:

• To compare Olink Explore 3072 and SomaScan v4 platforms using UK Biobank and Icelandic cohorts.
• To assess platform performance in detecting cis protein quantitative trait loci (pQTLs) across diverse ancestries.
• To evaluate how platform differences impact disease association studies.

Main Methods:

• Association studies on Olink Explore 3072 data from >50,000 UK Biobank participants across British/Irish, African, and South Asian ancestries.
• Comparison with SomaScan v4 data from 36,000 Icelandic individuals, including 1,514 with Olink data.
• Analysis of cis-pQTLs and genomic associations, stratified by ancestry.

Main Results:

• Modest correlation observed between Olink and SomaScan platforms.
• Similar absolute numbers of cis-pQTLs detected (Olink: 2,101; SomaScan: 2,120), but Olink showed a higher proportion of validated assays (72% vs. 43%).
• Significant differences in genomic associations for numerous proteins were identified between platforms, potentially influencing disease study conclusions.
• Leveraging diverse ancestries in UK Biobank enhanced novel association detection and genomic localization.

Conclusions:

• Both Olink and SomaScan are valuable high-throughput proteomics platforms, but exhibit distinct characteristics.
• Platform differences can influence research outcomes, necessitating careful consideration in integrated analyses.
• Utilizing diverse ancestries is crucial for robust proteogenomic association studies.
• Complementarity between platforms can enhance the depth and breadth of biological insights.