Protein S

Introduction

Protein S is a vitamin K-dependent plasma glycoprotein that plays a crucial role in the body’s natural anticoagulant system. It is synthesized primarily in the liver, as well as in endothelial cells, megakaryocytes, and Leydig cells. Its primary function is to act as a cofactor for activated Protein C (APC), a serine protease that inactivates coagulation factors Va and VIIIa, thereby limiting thrombin generation and preventing excessive blood clotting.

Biological Basis

The biological function of Protein S is intrinsically linked to the coagulation cascade. In its free form, Protein S binds to activated Protein C (APC), significantly enhancing APC’s ability to cleave and inactivate coagulation factors Va and VIIIa. This inactivation slows down the coagulation process, preventing the uncontrolled formation of blood clots. A significant portion of Protein S circulates in the plasma bound to C4b-binding protein (C4BP), an acute-phase reactant, which renders it inactive as an anticoagulant cofactor. Only the unbound, or “free,” Protein S is functionally active. The gene responsible for encoding Protein S isPROS1. Variations within PROS1can lead to altered Protein S levels or function.

Clinical Relevance

Measuring Protein S is clinically relevant for diagnosing and managing various coagulation disorders, particularly inherited or acquired thrombophilia, a condition characterized by an increased tendency to form blood clots. Low levels of functional or free Protein S can significantly increase an individual’s risk of venous thromboembolism (VTE), which includes deep vein thrombosis (DVT) and pulmonary embolism (PE). Protein S deficiency can be hereditary, often due to mutations in thePROS1gene, or acquired, caused by conditions such as liver disease, vitamin K deficiency, disseminated intravascular coagulation (DIC), pregnancy, or certain medications. Both quantitative (measuring total or free Protein S antigen) and functional assays are used to assess Protein S levels and activity.

Social Importance

Understanding and measuring Protein S levels has considerable social importance in public health and personalized medicine. Accurate diagnosis of Protein S deficiency allows for appropriate risk stratification and management strategies, including anticoagulant therapy, to prevent potentially life-threatening thrombotic events. This is particularly important for individuals with a family history of thrombosis or those experiencing recurrent clotting episodes. By identifying individuals at higher risk, healthcare providers can implement preventative measures, improve patient outcomes, and enhance quality of life. Furthermore, research into the genetic basis of Protein S levels contributes to a broader understanding of complex coagulation disorders and informs the development of targeted diagnostic and therapeutic approaches.

Methodological and Statistical Constraints

The interpretation of findings related to protein s is subject to several methodological and statistical limitations. A primary concern is the statistical power, as studies often require extremely large sample sizes, sometimes exceeding 35,000 individuals, to detect associations with small effect sizes that achieve genome-wide significance.^[1] Many studies, including those with pooled analyses involving thousands of subjects (e.g., 2,500-3,500 for various traits), may still have insufficient power to robustly identify less frequent variants or those with subtle effects, even if their impact is substantial.^[2] Furthermore, initial reports of genetic associations can be susceptible to the “winner’s curse,” where effect sizes are overestimated, potentially leading to inflated power estimates and hindering subsequent replication efforts.^[3] The extensive multiple testing inherent in genome-wide association studies also necessitates stringent significance thresholds, which can obscure true biological signals amidst statistical noise.^[3] While methods like genomic control and fixed-effects meta-analysis are employed to correct for inflation and combine results, the underlying assumptions of these statistical models, such as the identical distribution of SNP effects across the genome or accurate estimation of error covariance, can introduce biases if violated.^[2]

Phenotypic Definition and Accuracy

The accurate and consistent of protein s is crucial, yet challenges in phenotypic definition and data quality can impact the reliability of genetic associations. Systematically small differences in factors like sample DNA concentration, quality, handling procedures, or genotyping platforms can readily produce spurious effects that obscure true associations.^[4] Despite rigorous quality control protocols, infallible detection of incorrect genotype calls remains difficult, requiring a careful balance between stringent and lenient criteria for SNP exclusion.^[4]Furthermore, the variability in how protein s might be indirectly assessed or collected across different cohorts, or even the absence of certain measures in specific datasets, can introduce heterogeneity that complicates meta-analysis and the overall interpretation of genetic influences.^[2] Such technical artifacts and errors contribute to the overall estimation error, potentially biasing the observed effect sizes of genetic variants.^[5]

Generalizability and Unaccounted Confounding

The generalizability of genetic findings for protein s is often limited by the demographic characteristics of the study populations. Many large-scale genetic analyses primarily involve individuals of European ancestry, which restricts the direct applicability of these findings to populations with different ancestral backgrounds.^[5]This ancestral bias can lead to an incomplete understanding of how genetic variants associated with protein s levels might vary in frequency or effect across diverse global populations. Population stratification, where differences in allele frequencies correlate with varying protein s levels due to ancestral substructure rather than direct genetic causation, poses a significant confounder, even when statistical adjustments for genetic ancestry are applied.^[6]Beyond genetic factors, the influence of unmeasured environmental factors, lifestyle choices, or complex gene-environment interactions on protein s levels is often not fully captured or modeled in current genetic studies. This omission contributes to the “missing heritability” phenomenon, where a substantial proportion of the heritable variation in protein s remains unexplained by identified genetic variants, indicating a gap in our comprehensive understanding of its genetic architecture.

Variants

Genetic variants, or single nucleotide polymorphisms (SNPs), can significantly influence gene function, protein expression, and ultimately, an individual’s health and susceptibility to various conditions. The variants discussed here are associated with genes involved in fundamental cellular processes, from gene regulation and protein folding to neuronal development and metabolic pathways, thereby potentially affecting circulating protein levels and related traits.

The genomic region encompassing SUDS3 and LINC02460, along with the gene PCGF3, highlights critical aspects of gene regulation. SUDS3 (Suppressor of Defects in S3) is an integral component of the histone deacetylase (HDAC) complex, which modifies chromatin structure to control gene expression. A variant such as rs11613092 could subtly alter SUDS3 function, thereby impacting the epigenetic landscape and the transcription of many genes, leading to changes in protein production. Similarly, LINC02460 is a long intergenic non-coding RNA, known to regulate gene expression through diverse mechanisms, including acting as a scaffold for protein complexes. PCGF3 (Polycomb Group RING Finger 3) plays a vital role in the Polycomb repressive complex 1 (PRC1), a major regulator of gene silencing and cellular memory. The variant rs4234853 , located in proximity to PCGF3, may affect the stability or activity of PRC1, leading to altered gene silencing patterns and consequently, modified protein levels crucial for development and cell differentiation.^[7] These epigenetic modifications are essential for maintaining cellular identity and responding to environmental signals, and their precise regulation is paramount for overall health, with potential implications for various complex traits.^[2] Further variants implicate genes with roles in cellular signaling, development, and protein quality control. LINC02315 is another long intergenic non-coding RNA, whose expression or function, potentially influenced by rs1959947 , could modulate the levels of numerous proteins by affecting chromatin remodeling or transcriptional interference. The genomic locus for NKX1-2 and LHPP is significant for its involvement in developmental and signaling pathways. NKX1-2 is a homeobox gene critical for nervous system development, while LHPP(Phospholysine Phosphohistidine Inorganic Pyrophosphate Phosphatase) is an enzyme that regulates protein activity through histidine phosphorylation. The variantrs2459210 might alter the expression or function of these genes, potentially impacting protein phosphorylation states and a broad spectrum of cellular processes, including neuronal function and metabolism.^[8] Moreover, DNAJC6 (DnaJ Heat Shock Protein Family (Hsp40) Member C6) encodes a co-chaperone protein essential for correct protein folding and intracellular trafficking, particularly in clathrin-mediated endocytosis. Variations such as rs1413885 in DNAJC6 could compromise its chaperone activity, resulting in misfolded proteins or defects in endocytosis, thereby disrupting cellular communication and nutrient uptake, and ultimately affecting the cellular protein landscape.^[7] Such disruptions have been linked to neurodegenerative conditions.

Other variants affect genes involved in RNA processing, neuronal guidance, and mitochondrial function. The region containing LINC02383 and RN7SL691P underscores the diverse functions of non-coding RNAs. LINC02383 is a long intergenic non-coding RNA, while RN7SL691P is a pseudogene related to 7SL RNA, a component of the Signal Recognition Particle (SRP) crucial for targeting proteins to the endoplasmic reticulum for synthesis and secretion. The variant rs13130255 could influence the expression or stability of these non-coding RNAs, thereby impacting the efficiency of protein synthesis and trafficking. The genes HMG20A (High Mobility Group 20A) and LINGO1(Leucine Rich Repeat And Ig Domain Containing Nogo Receptor Interacting Protein 1) are central to neuronal processes.HMG20A is a chromatin-associated protein involved in regulating gene transcription during neuronal development, and LINGO1 acts as an inhibitor of neuronal differentiation and axon regeneration.^[2] A variant like rs2137111 in this genomic region could disrupt the intricate balance of neuronal growth and plasticity, potentially altering protein levels associated with brain function and structure. Finally, the EPHB1 and SDHBP1 locus is important for cell-cell communication and energy metabolism. EPHB1 (Ephrin Receptor B1) is a receptor tyrosine kinase involved in various developmental processes, including axon guidance, while SDHBP1(Succinate Dehydrogenase Complex Assembly Factor 1) is crucial for the assembly of succinate dehydrogenase, a key enzyme in mitochondrial respiration.^[8] The rs1401543 variant could therefore impact either neuronal signaling or mitochondrial energy production, both of which are fundamental processes influencing the overall protein expression and activity profile within cells, with broad implications for metabolic and neurological health.

Key Variants

RS ID	Gene	Related Traits
rs11613092	SUDS3 - LINC02460	Alzheimer disease protein s
rs4234853	PCGF3	protein s
rs1959947	LINC02315	protein s
rs2459210	NKX1-2 - LHPP	protein s
rs1413885	DNAJC6	protein s
rs13130255	LINC02383 - RN7SL691P	protein s
rs2137111	HMG20A - LINGO1	protein s
rs1401543	EPHB1 - SDHBP1	protein s

High-Throughput Plasma Proteome Analysis

Advanced techniques for comprehensive plasma protein analysis are fundamental in diagnosing conditions characterized by altered protein expression. The multiplexed, aptamer-based SOMAscan assay, for instance, provides a broad profiling of thousands of plasma proteins and protein complexes, including extracellular and intracellular components, as well as soluble domains of membrane-associated proteins.^[9]This method offers an extended lower limit of detectable protein abundance compared to conventional immunoassays, enabling a more thorough assessment of the plasma proteome. The selection of proteins on this platform is guided by their suspected involvement in the pathophysiology of human disease and their wide range of molecular functions, making it a valuable tool for identifying potential diagnostic biomarkers.^[9]

Reliability and Quantitation of Plasma Biomarkers

The accurate diagnosis of various conditions relies critically on the reproducible quantitation of specific biomarkers in plasma. Biochemical assays are routinely employed to measure concentrations of key inflammatory and metabolic markers, such as C-reactive protein (CRP), interleukin-6, soluble intracellular adhesion molecule-1, monocyte chemoattractant protein-1 (MCP1), and myeloperoxidase.^[7] The reliability of these measurements is paramount for clinical utility, with studies demonstrating good intra-assay coefficients of variation for numerous biomarkers, including CD40 ligand (4.4%), interleukin-6 (3.1%), and tumor necrosis factor receptor-2 (2.3%).^[7]Inter-assay reproducibility is also rigorously assessed for markers like brain natriuretic peptide (12.2%) and n-terminal-atrial natriuretic peptide (12.7%), ensuring consistency in diagnostic evaluations, further supported by high concordance rates such as a Kappa statistic of 0.95 for CRP samples run in duplicate.^[7]

Clinical Interpretation of Protein Data

The clinical utility of protein data in diagnosis stems from its capacity to reveal insights into disease states and physiological functions, such as vitamin K metabolism or systemic inflammation. For example, measurements of % undercarboxylated osteocalcin and various inflammatory markers contribute to understanding underlying biological processes relevant to disease.^[7]While specific diagnostic criteria or physical examination findings are not detailed, the identification of proteins with known roles in human disease pathophysiology through comprehensive profiling enables clinicians to discern deviations from healthy physiological ranges. This facilitates the identification of potential disease markers or indicators of altered biological function, guiding further clinical investigation and informing patient management strategies.^[7]

Genetic and Epigenetic Determinants of Protein Abundance

The abundance of proteins, including protein S, is fundamentally shaped by genetic and epigenetic factors, which orchestrate gene regulation from transcription through translation. Genetic variants, often identified through genome-wide association studies (GWAS), can act as protein quantitative trait loci (pQTLs) that directly influence the levels of specific proteins in the blood plasma.^[10] These genetic differences can impact the expression of genes encoding transcription factors or components of cell signaling pathways, thereby exerting broad control over the proteome.^[11] For instance, specific quantitative trait loci (QTLs) have been identified that underlie proteome variation in human lymphoblastoid cells, highlighting how genetic architecture contributes to the diversity of protein levels across individuals.^[12]Beyond sequence variation, epigenetic modifications, such as DNA methylation, also play a crucial role in regulating gene expression and, consequently, protein levels, demonstrating a complex interplay between the genome and epigenome in shaping the plasma proteome.^[13]This intricate regulatory layer ensures that cellular protein concentrations are maintained within a functional range, adapting to internal and external cues. The genetic control of protein abundance can be quite specific, influencing the levels of a single protein or a cluster of functionally related proteins. These regulatory mechanisms include the binding of transcription factors to specific DNA sequences to modulate gene transcription, as well as feedback loops that fine-tune the expression levels of genes based on the cellular demand for their protein products. Understanding these genetic and epigenetic influences is critical for deciphering the baseline variability of protein S levels within a population and for identifying individuals with predispositions to altered protein S concentrations due to their genetic makeup.

Cellular Signaling and Post-Translational Control

Protein S levels are also dynamically regulated through complex cellular signaling pathways and diverse post-translational modifications (PTMs) that affect protein activity, stability, and localization. Receptor activation at the cell surface initiates intracellular signaling cascades, involving a series of protein-protein interactions and enzymatic modifications, such as phosphorylation, which can rapidly alter the functional state of numerous downstream proteins. These cascades can ultimately lead to the activation or inhibition of transcription factors, thereby influencing the synthesis rates of proteins, including protein S.^[11] For example, specific genetic variants influencing transcription factor and cell signaling protein levels have been identified, underscoring the genetic basis of these dynamic regulatory networks.

Post-translational modifications are critical regulatory mechanisms that provide another layer of control over protein function without altering gene expression. These modifications, which include phosphorylation, glycosylation, ubiquitination, and cleavage, can dramatically change a protein’s conformation, its ability to interact with other molecules, or its half-life. Allosteric control, where binding of a molecule at one site on a protein affects the activity at another site, represents a rapid and reversible regulatory mechanism integral to many signaling and metabolic pathways. The balance of these modifications, often influenced by feedback loops within signaling networks, ensures precise control over the activity and availability of protein S, allowing cells to respond promptly to physiological changes.

Metabolic Pathway Regulation

The levels and activities of proteins, including protein S, are intimately linked with metabolic pathways, which govern energy metabolism, biosynthesis, and catabolism within the body. Genetic variations can significantly influence human metabolism, affecting the flux through various metabolic pathways and consequently altering the demand for or production of specific proteins.^[14] For instance, common genetic variation near MC4Rhas been associated with waist circumference and insulin resistance, highlighting a link between genetic predisposition, metabolic state, and potentially the proteins involved in these processes.^[15] Similarly, variation in MLXIPL is associated with plasma triglycerides, indicating genetic control over lipid metabolism and the proteins that regulate it.^[15] Metabolic regulation involves intricate mechanisms that control the rate of biochemical reactions, ensuring that energy production and nutrient utilization are balanced according to cellular needs. This includes the regulation of enzyme activity through allosteric control, covalent modification, and changes in enzyme concentration via gene regulation. The FTOobesity variant, for example, impacts metabolic circuitry by influencing adipocyte browning, a process critical for energy expenditure, illustrating how genetic variations can reshape metabolic pathways and the proteins that mediate them.^[16] Changes in metabolic flux directly influence the availability of precursors for protein synthesis and the efficiency of protein degradation pathways, thereby exerting profound control over overall protein abundance.

Inter-Pathway Crosstalk and Disease Relevance

Biological systems operate through highly interconnected networks where various pathways constantly communicate and influence one another, a phenomenon known as pathway crosstalk. This systems-level integration ensures coordinated cellular responses and maintains homeostasis, with hierarchical regulation dictating the overall cellular state and emergent properties. Genetic risk factors for diseases often manifest their effects through such integrated networks, connecting genetic variation to disease endpoints via alterations in the human blood plasma proteome.^[10]For instance, specific transcription factor binding site patterns have been leveraged to link diabetes risk loci to underlying disease mechanisms, demonstrating how genetic variations in regulatory elements can impact multiple pathways simultaneously.^[16]Dysregulation within these complex pathways and their crosstalk is a hallmark of many diseases. For example, aptamer-based proteomic profiling has revealed novel candidate biomarkers and pathways implicated in cardiovascular disease, indicating that altered protein levels reflect underlying pathological processes.^[17]In disease states, compensatory mechanisms may activate alternative pathways to mitigate the initial insult, but these can also contribute to disease progression over time. Identifying key proteins and pathways that are dysregulated offers promising therapeutic targets, as modulating their activity can restore balance to the system. Understanding the genetic architecture of gene expression in specific tissues, such as the human liver, provides insights into how disease-relevant proteins are controlled and how their aberrant regulation contributes to pathophysiology.^[18]

Clinical Relevance

The of plasma proteins offers significant clinical utility across various domains, from diagnostics and risk assessment to understanding disease mechanisms and monitoring patient outcomes. Modern high-throughput platforms, such as the multiplexed, aptamer-based SOMAscan assay, enable the quantification of thousands of proteins, including both extracellular and intracellular components, with a focus on those implicated in human disease pathophysiology.^[9] The reliability and reproducibility of these protein measurements, demonstrated through studies reporting good coefficients of variation for various biomarkers, underpin their potential for widespread clinical application.^[7]

Diagnostic Utility and Risk Stratification Approaches

Plasma protein measurements serve as crucial tools for diagnostic assessment and stratifying individuals based on their disease risk. The ability to measure a vast array of proteins, encompassing diverse molecular functions, allows for a comprehensive assessment of physiological states and disease-associated phenotypes.^[9] For instance, studies employing these methods in large cohorts, such as the UK Biobank with its extensive plasma protein trait data, contribute to identifying protein signatures associated with various conditions.^[8]Specific biomarker traits, including inflammatory markers like CRP, interleukin-6, and myeloperoxidase, alongside indicators such as brain natriuretic peptide and aspartate aminotransferase, have demonstrated good reproducibility, making them valuable for informing diagnostic considerations and identifying individuals at elevated risk for particular health outcomes.^[7]This comprehensive profiling facilitates a more granular understanding of patient health status, moving towards personalized medicine approaches in disease prevention and early detection.

Prognostic Value and Monitoring Disease Trajectories

The dynamic nature of plasma protein levels provides valuable insights into disease progression, treatment response, and long-term patient outcomes, establishing their role as prognostic indicators. Longitudinal studies, involving the contemporaneous assay of baseline and two-year samples, highlight the potential for monitoring changes in protein concentrations over time, which can reflect disease activity or therapeutic efficacy.^[9]The consistent and reproducible of biomarker traits, as observed in studies like the Framingham Heart Study, supports their application in tracking disease trajectories and predicting future clinical events.^[7]Such monitoring strategies can guide adjustments to treatment regimens, assess the effectiveness of interventions, and ultimately improve patient management by providing objective measures of disease course and response to therapy.

Understanding Disease Pathophysiology and Comorbidity Associations

The extensive profiling of plasma proteins is instrumental in elucidating the underlying pathophysiology of human diseases and identifying associations with related conditions, complications, and overlapping phenotypes. The selection of proteins for advanced platforms is deliberately biased towards those “suspected to be involved in the pathophysiology of human disease,” covering a wide range of molecular functions.^[9] Large-scale genomic studies, which analyze plasma protein traits in conjunction with genetic data, help uncover the genetic determinants of protein levels, thereby revealing fundamental biological pathways and their connections to various health conditions.^[8]This allows for a deeper understanding of complex disease etiology, the identification of shared molecular mechanisms between comorbidities, and the elucidation of syndromic presentations, offering new avenues for therapeutic targeting and integrated patient care.

Frequently Asked Questions About Protein S

These questions address the most important and specific aspects of protein s based on current genetic research.

---

1. My mom had blood clots; will I get them too?

Yes, if your mother had a Protein S deficiency, there’s a chance you could inherit it. This deficiency is often hereditary, caused by variations in thePROS1gene, which significantly increases your risk of developing blood clots like deep vein thrombosis or pulmonary embolism. Measuring your Protein S levels can help assess your personal risk.

2. Does being pregnant increase my risk for blood clots?

Yes, pregnancy is one of the conditions that can cause an acquired Protein S deficiency. This means your body might have lower levels of active Protein S during pregnancy, potentially increasing your risk of venous thromboembolism. Your doctor might monitor your levels if you have other risk factors or a family history.

3. Can my daily medicines affect my clotting risk?

It depends on the medication. Certain medications are mentioned as potential causes of acquired Protein S deficiency, which could increase your risk of blood clots. It’s important to discuss all your medications with your doctor, especially if you have a history of clotting or are being tested for Protein S levels.

4. Does my diet, especially vitamin K, affect my risk for clots?

Yes, your diet, specifically your vitamin K intake, can play a role. Protein S is a vitamin K-dependent protein, and a deficiency in vitamin K can lead to lower levels of functional Protein S. This acquired deficiency can increase your tendency to form blood clots.

5. If I’ve had a DVT, what does a Protein S test tell me?

If you’ve had a deep vein thrombosis (DVT), a Protein S helps determine if a deficiency in this protein contributed to it. Low levels of functional or free Protein S significantly increase your risk of such events, and the test can help diagnose inherited or acquired thrombophilia, guiding future prevention.

6. If I’m at risk for clots, what can I do?

If you’re identified as being at higher risk due to Protein S deficiency, your doctor might recommend specific strategies. These often include anticoagulant therapy to prevent potentially life-threatening thrombotic events. Understanding your risk allows for personalized preventative measures and improved quality of life.

7. Does my family’s background affect my clotting risk?

Research on genetic factors for Protein S levels has primarily focused on individuals of European ancestry. This means that genetic risks and their frequencies might differ in populations with other ancestral backgrounds, and specific findings might not directly apply to you if you are not of European descent.

8. How accurate are those genetic tests for clotting risk?

Genetic studies for Protein S can face challenges in accuracy. Factors like how samples are handled, DNA quality, and the genotyping platforms used can create small differences that might obscure true genetic associations. While quality control is rigorous, detecting all incorrect genotype calls is difficult, meaning tests might not always capture the full picture perfectly.

9. Why do some people get blood clots easily, but others don’t?

The tendency to form blood clots, or thrombophilia, can vary significantly between people. Some individuals have inherited conditions, often due to variations in genes like PROS1, that lead to low or dysfunctional Protein S. Others might develop acquired deficiencies due to conditions like pregnancy or liver disease, making them more prone to clotting.

10. Does my liver health affect my risk for blood clots?

Yes, your liver health is crucial for Protein S. Protein S is primarily synthesized in the liver, so conditions like liver disease can lead to an acquired Protein S deficiency. This reduction in functional Protein S can increase your overall risk of developing blood clots.

---

This FAQ was automatically generated based on current genetic research and may be updated as new information becomes available.

Disclaimer: This information is for educational purposes only and should not be used as a substitute for professional medical advice. Always consult with a healthcare provider for personalized medical guidance.

References

[1] Xing, C. et al. “A weighted false discovery rate control procedure reveals alleles at FOXA2 that influence fasting glucose levels.”American Journal of Human Genetics, vol. 86, no. 2, 2010, pp. 210-18.

[2] Gialluisi, Alessandro et al. “Genome-wide association scan identifies new variants associated with a cognitive predictor of dyslexia.” Translational Psychiatry, vol. 9, no. 1, 2019, p. 57.

[3] Liu, J. Z. et al. “Genome-wide association study of height and body mass index in Australian twin families.”Twin Research and Human Genetics, vol. 13, no. 2, 2010, pp. 119-30.

[4] Wellcome Trust Case Control Consortium. “Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls.” Nature, vol. 447, no. 7145, 2007, pp. 661-78.

[5] Turley, P. et al. “Multi-trait analysis of genome-wide association summary statistics using MTAG.” Nature Genetics, vol. 50, no. 2, 2018, pp. 224-34.

[6] Plenge, R. M. et al. “Two independent alleles at 6q23 associated with risk of rheumatoid arthritis.”Nature Genetics, vol. 39, no. 12, 2007, pp. 1477-82.

[7] Benjamin, E. J., et al. “Genome-wide association with select biomarker traits in the Framingham Heart Study.” BMC Medical Genetics, vol. 8, no. Suppl 1, 2007, p. S11.

[8] Loya, Hagai, et al. “A scalable variational inference approach for increased mixed-model association power.” Nature Genetics, vol. 56, 2024, pp. 297–308.

[9] Sun, B. B., et al. “Genomic atlas of the human plasma proteome.” Nature, 2018.

[10] Suhre K et al. “Connecting genetic risk to disease end points through the human blood plasma proteome.”Nat Commun, 2017.

[11] Hause, R. J. et al. “Identification and validation of genetic variants that influence transcription factor and cell signaling protein levels.” Am. J. Hum. Genet., vol. 95, 2014, pp. 194–208.

[12] Garge, N. et al. “Identification of quantitative trait loci underlying proteome variation in human lymphoblastoid cells.” Mol. Cell. Proteomics, vol. 9, 2010, pp. 1383–1399.

[13] Petersen, A. -K. K. et al. “Epigenetics meets metabolomics: an epigenome-wide association study with blood serum metabolic traits.” Hum. Mol. Genet., 2014.

[14] Illig, T. et al. “A genome-wide perspective of genetic variation in human metabolism.” Nat. Genet., vol. 42, 2010, pp. 137–141.

[15] Kooner, J.S. et al. “Common genetic variation near MC4R is associated with waist cir-cumference and insulin resistance.”Nat. Genet., vol. 40, 2008, pp. 716–718.

[16] Claussnitzer, M. et al. “FTO obesity variant circuitry and adipocyte browning in humans.”N. Engl. J. Med., vol. 373, 2015, pp. 895–907.

[17] Ngo, D. et al. “Aptamer-based proteomic profiling reveals novel candidate biomarkers and pathways in cardiovascular disease.”Circulation, vol. 134, 2016, pp. 270–285.

[18] Schadt, E.E. et al. “Mapping the genetic architecture of gene expression in human liver.” PLoS Biol., vol. 6, 2008, p. e107.