Machine learning cross-platform proteomic imputation enables protein quality scoring and replication of epidemiological associations

ABSTRACT High-throughput affinity-based proteomics has advanced biomedical research, yet fundamental, persistent discordance between mainstream platforms (SomaScan and Olink) routinely undermines the replication of findings. This platform-driven non-replication complicates downstream biological validation and biomarker prioritization. Here, we develop a machine learning-based framework for cross-platform protein value imputation to resolve this translational bottleneck. Using paired proteomic data measured by both SomaScan and Olink from 5,325 participants of the Multi-Ethnic Study of Atherosclerosis, we developed models to impute cross-platform measurements and applied them to two independent and demographically distinct cohorts (Cardiovascular Health Study [N=3,171] and UK Biobank [UKB; N=41,405]) for external validation. Our bi-directional model 1) established an imputation performance-based protein fidelity index, validated against gold-standard measurements from Atherosclerosis Risk in Communities study (N=101) and Nurses’ Health Study (N=54), 2) enabled imputation of platform-exclusive protein measurements, and 3) facilitated calibration of overlapping proteins. We demonstrate the utility of this framework through three applications: 1) fidelity-informed analyses enhanced the replication of biomarker discovery, 2) recovery of SomaScan signals that were previously inaccessible in UKB’s original Olink measurements, and 3) improved replication performance for overlapping proteins. Our study offers a translational roadmap that allows researchers to achieve reliable epidemiological replication, target specific assays for future optimization, and prioritize biological signal over platform noise. Competing Interest Statement P.N. reports research grants from Allelica, Amgen, Apple, Boston Scientific, Cleerly, Genentech / Roche, Ionis, Novartis, and Silence Therapeutics, personal fees from AIRNA, Allelica, Apple, AstraZeneca, Bain Capital, Blackstone Life Sciences, Bristol Myers Squibb, Creative Education Concepts, CRISPR Therapeutics, Eli Lilly & Co, Esperion Therapeutics, Foresite Capital, Foresite Labs, Genentech / Roche, GV, HeartFlow, Incyte, Magnet Biomedicine, Merck, Novartis, Novo Nordisk, TenSixteen Bio, and Tourmaline Bio, equity in Bolt, Candela, Mercury, MyOme, Parameter Health, Preciseli, and TenSixteen Bio, royalties from Recora for intensive cardiac rehabilitation, and spousal employment at Vertex Pharmaceuticals, all unrelated to the present work. C.M.B. reports receiving grant / research support from: Abbott Diagnostic, Akcea, Amgen, Arrowhead, Eli Lilly, Ionis, Merck, New Amsterdam, Novartis, Novo Nordisk, Roche Diagnostic (All paid to institution, not individual): Consultant: 89Bio, Abbott Diagnostics, Amgen, Arrowhead, Astra Zeneca, Denka Seiken,* Esperion, Genentech, HeartFlow, Ionis, Eli Lilly,* Merck,* New Amsterdam, Novartis, Novo Nordisk, Roche Diagnostic. *Significant where noted (>$10,000); remainder modest (<$10,000). L.M.R. is a consultant for the NHLBI TOPMed program's Administrative Coordinating Center (through Westat). A.P. reports employment at Google Ventures.