SEARCH · PubMed Health

Results for “Olink platform”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

14 recordsLinked to original sources

PubMed2026

Candidate biomarker identification for blood stasis syndrome among coronary artery disease patients using the Olink proteomics platform.

OBJECTIVE: To identify candidate biomarkers of blood stasis syndrome (BSS) associated with coronary artery disease (CAD) and explore the underlying inflammatory mechanisms. METHODS: Using the Olink Target 96 Inflammation panel, we identified plasma proteins in a group of 88 patients comprised of healthy controls (HCs), those with CAD and BSS (CAD-BSS), those with CAD without BSS (CAD-non-BSS), and those with BSS without CAD (non-CAD-BSS) (n = 22 in each group). Protein molecules that were specifically expressed in CAD or BSS were identified by differential expression analyses. Subsequently, potential protein biomarkers were identified using least absolute shrinkage and selection operator regression to enable CAD and BSS differentiation. The potential functional mechanisms of identified proteins were then determined by Gene Ontology enrichment and Kyoto Encyclopedia of Genes and Genomes pathway analyses. RESULTS: Patients with CAD had 31/92 upregulated and 4/92 downregulated proteins compared with those without. Chemokine (C-C motif) ligand 11 (CCL11), CUB domain-containing protein 1, hepatocyte growth factor, sirtuin 2 (SIRT2), eukaryotic translation initiation factor 4E-binding protein 1 (4E-BP1), CCL25, and tumor necrosis factor (TNF) showed the strongest upregulation (all P <0.0001). Patients with BSS had 8/92 downregulated proteins, specifically CCL28, CCL11, cystatin D, STAM-binding protein, 4E-BP1, matrix metalloproteinase-10, SIRT2, and monocyte chemotactic protein 4, compared with those without (all P < 0.05). The CAD-BSS group had one interleukin-17 (IL-17) upregulated and 10/92 downregulated proteins compared with the CAD-non-BSS group. When compared with the non-CAD-BSS group, the CAD-BSS group had 8 upregulated proteins but only 2 downregulated proteins, namely interleukin-10 receptor subunit alpha (IL-10RA) and TNF-related activation-induced cytokine (both P < 0.05). Totally 10 proteins were identified as potential candidate biomarkers of BSS in CAD patients. After least absolute shrinkage and selection operator regression analysis, two proteins that distinguished between BSS and non-BSS individuals among CAD patients were identified (SIRT2 and 4E-BP1). These proteins are primarily associated with the mechanistic target of rapamycin signaling pathway, which regulates inflammation and oxidative stress. CONCLUSIONS: Results suggest that the inflammatory response and mechanistic target of rapamycin signaling pathway participate in CAD and BSS development, and that SIRT2 and 4E-BP1 are prospective protein biomarkers for patients with CAD and BSS.

Humans↗

PubMed2026

Proteomics as a theranostic compass in BCR::ABL1-negative myeloproliferative neoplasms: Integrating biomarker discovery with therapeutic stratification.

Classic BCR::ABL1-negative myeloproliferative neoplasms (MPNs)-polycythaemia vera, essential thrombocythaemia, and primary myelofibrosis-are clonal haematopoietic stem cell disorders with marked heterogeneity in clinical phenotype, disease trajectory, and therapeutic response. Genomic stratification by driver and cooperating mutations only partially accounts for this variability, leaving gaps in predicting thrombotic risk, fibrotic progression, leukaemic transformation, and treatment benefit. Proteomics bridges this gap by providing function-proximal readouts of protein abundance, post-translational modifications, pathway activity, and intercellular signalling that genomics and transcriptomics cannot capture, positioning it as a theranostic platform in which the same molecular readouts simultaneously inform diagnostic stratification and therapeutic decision-making. We propose a five-stage translational framework spanning from discovery-scale mass spectrometry and affinity-based plasma profiling to targeted validation, multicentre standardisation, and machine learning-integrated clinical panels. Proteomic evidence is synthesised across the following four disease axes: clonal fitness in haematopoietic stem and progenitor cells; bone marrow microenvironmental remodelling and fibrosis; chronic inflammation and thrombosis; and leukaemic transformation. We further describe how phosphoproteomics reveals resistance mechanisms to JAK inhibitors, including AXL-MAPK bypass and PP2A-autophagy-mediated tolerance, and how protein-level biomarkers (BCL2-BCL-XL, RAS-ERK, CAMK2G, and ROCK1/2) can guide individualised therapeutic selection. Affinity-based platforms (Olink PEA and SomaScan) and spatially resolved technologies (CODEX and single-cell proteomics) complement discovery proteomics. At present, however, this evidence base is constrained by small and heterogeneous cohorts, limited cross-platform reproducibility, and a scarcity of independent external validation for candidate protein panels. Realising this vision will require multicentre standardisation, analytically validated panel assays, and prospective clinical studies that translate molecular findings into decision-grade tools for patients with MPNs.

Humans↗

PubMed2026

Identification of Immune Response-Related Proteomic Biomarkers in Moyamoya Disease Using Serum Olink Proteomics.

Moyamoya disease, a rare chronic cerebrovascular disorder, requires invasive digital subtraction angiography (DSA) for diagnosis. This study employed high-throughput proteomics to identify plasma biomarkers for Moyamoya disease diagnosis. We conducted immunopanel analysis using the Olink platform to evaluate 92 immune-related proteins in plasma samples from 88 Moyamoya disease patients and 88 healthy controls. Key proteins were identified through differential expression analysis, GO, and KEGG enrichment analysis. A diagnostic model was constructed using LASSO regression, Boruta algorithm, and machine learning models including random forest and XGBoost. Validation of these proteins was performed using GEO external data sets, followed by prediction of potential therapeutic drugs and molecular docking validation through pharmacogenomic databases. A total of 44 differentially expressed proteins were identified through the Olink immunopanel, with 12 downregulated and 32 upregulated. GO and KEGG analyses revealed significant enrichment of these proteins in innate immune responses and signaling pathways such as NF-kB and MAPK. Through LASSO, random forest, and protein under-area analysis, four potential biomarkers for Moyamoya disease (MGMT, SIT1, PRDX1, TRAF2) were identified. A diagnostic model using these proteins showed the highest AUC value with the XGBoost model. Additionally, TRAF2 and PRDX1 exhibited significant expression differences in Moyamoya disease patients within the GEO data set. Our study revealed the immune landscape of Moyamoya disease, identified four biomarkers, and established a variety of diagnostic models.

Humans↗

PubMed2026

Proteomic patterns according to ejection fraction: an EMPEROR-programme analysis.

AIMS: Left ventricular ejection fraction (LVEF) has been incorporated as an inclusion criterion in HF trials. Patient's characteristics, event risk, and treatment response vary according to LVEF. A better understanding of the biological processes across LVEF is warranted. To study proteomic biomarker expression across LVEF using data from the EMPEROR-Programme. METHODS: Two thousand two hundred and fifty-four patients who had proteomic measurements available using 1134 proteins overlapping between the Explore 1536 and 3072 Olink® platforms were included. Main analyses were performed within the EMPEROR-Preserved dataset due to differences in entry criteria between EMPEROR-Preserved and EMPEROR-Reduced with higher entry N-terminal pro B-type natriuretic peptide (NT-proBNP) levels that varied by LVEF cut-offs in the latter. Protein concentrations were compared using ordinal logistic regression across LVEF categories: 41%-49%, 50%-59%, and ≥60%. The resulting β-coefficient indicates the change in the log-odds for the outcome of being in a lower LVEF category for every NPX unit in log2 scale. Analyses were adjusted for covariates and a false-discovery-rate (FDR) correction was applied. RESULTS: A total of 297 proteins exhibited a trend of expression across LVEF categories in EMPEROR-Preserved after adjustment for potential confounders and correction for test multiplicity. Of these, the top 10 proteins were: NT-pro BNP (β = 0.18, 95% CI 0.09-0.27), Wnt inhibitory factor-1 (β = 0.40, 95% CI 0.19-0.61), sialomucin core protein 24 (β = 0.48, 95% CI 0.22-0.74), phospholipid transfer protein (β = 0.38, 95% CI 0.17-0.59), natriuretic peptides B (β = 0.13, 95% CI 0.06-0.20), intercellular adhesion molecule 5 (β = 0.31, 95% CI 0.14-0.49), neural cell adhesion molecule 2 (β = 0.45, 95% CI 0.19-0.70), neural cell adhesion molecule L1-like protein (β = 0.45, 95% CI 0.19-0.71), interactor protein for cytohesin exchange factors 1 (β = 0.12, 95% CI 0.05-0.19), and 3-ketoacyl-CoA thiolase, peroxisomal (β = 0.17, 95% CI 0.07-0.26). The correlation between these proteins and LVEF was generally weak (Rho ≤0.2). CONCLUSIONS: Within EMPEROR-Preserved, the top differentially expressed circulating proteins suggest that pathways related to natriuretic peptides, cell-adhesion, and clonal haematopoiesis are overexpressed at mildly-reduced ejection fraction, but none of the proteins passed the 5%FDR cut-off, and the correlation between circulating proteins and LVEF was weak. These findings suggest that circulating proteins may not be a good discriminant of ejection fraction.

Humans↗

PubMed2026

Proteomic Profiling Captures Residual Cardiovascular Risk Beyond the PREVENT Model in Individuals With Cardiovascular-Kidney-Metabolic Syndrome Stages 2-3.

BACKGROUND: Cardiovascular-kidney-metabolic (CKM) syndrome reflects complex pathobiological interactions among metabolic disorders, kidney injury, and cardiovascular disease (CVD). Stages 2 and 3 represent critical phases of disease progression characterised by high pathological heterogeneity. This study aimed to develop a CVD protein risk score (PRS) for this population and evaluate its incremental predictive value over the PREVENT model. METHODS: This study included 24 017 participants with CKM Stages 2-3 from the UK Biobank. Using 2923 plasma proteins measured via the Olink platform, a PRS was developed in a training set (n = 19 218) using the LASSO method. In the validation set (n = 4799), the incremental predictive performance of this score over the PREVENT model was assessed using Harrell's C-statistic, net reclassification improvement (NRI) and integrated discrimination improvement (IDI). RESULTS: A risk score comprising 63 proteins was constructed, primarily reflecting inflammation, kidney injury and matrix remodelling. Key proteins included growth differentiation factor 15 (GDF15), hepatitis A virus cellular receptor 1 (HAVCR1), matrix metallopeptidase 12 (MMP12) and NT-proBNP. In the validation set, after adjusting for PREVENT risk factors, individuals in the high PRS group had a 2.56-fold higher risk of CVD compared to those in the low score group (HR: 2.56, 95% CI: 1.96-3.37). Integrating the score into the PREVENT model improved the C-statistic by 0.034 (0.672-0.706) and achieved a 10-year NRI of 15.8% (95% CI: 9.5%-20.9%) and an IDI of 2.2% (95% CI: 1.3%-3.3%). CONCLUSION: Combining the PREVENT model with the PRS developed in this study enhances the prediction of future CVD events in the CKM Stages 2-3 population. This approach facilitates the capture of residual risk and supports precision risk stratification and management for this high-risk group.

Humans↗

PubMed2026

Comparison of endothelin-1 levels in plasma from human coronary arteries measured by enzyme linked immunosorbent assay and Olink high-throughput proteomics platform.

Endothelin-1 (ET-1) antagonists are increasingly being approved for new treatments for cardiovascular disease, where elevated ET-1 levels contribute to increased vasoconstriction. Further therapeutic targets, including coronary artery disease, are under investigation. The Olink Explore 3072 Proximity Extension Assay platform enables multiplexed high-throughput measurement of ~3000 plasma proteins, from minimal (≤6 µL) sample volumes. However, it is not known if the two oligonucleotide-tagged antibodies raised against preproET-11-212, used in this Olink assay, specifically measure biologically active ET-1 or the other inactive EDN1-encoded peptides, also secreted by human endothelial cells. Paired plasma samples from 29 patients with coronary artery disease were obtained, using a specialised intra-coronary sampling catheter, designed to obtain site specific biochemical information from within coronary arteries. We compared ET-1 concentrations measured with an ET-1 specific ELISA, demonstrated to have no cross-reactivity with other EDN1-encoded peptides versus values obtained using Olink Explore platform. Olink-measured ET-1 correlated significantly with ELISA-derived ET-1 levels (r = 0.53, p = 0.003), and Olink values predicted ELISA results. Olink ET-1 concentrations also correlated with ETB receptor levels (r = 0.40, p < 0.05). These findings indicate that the Olink Explore platform can detect relative changes in biologically active ET-1, supporting its use as a biomarker tool in clinical and translational studies.

Humans↗

PubMed2026

Serum Olink Proteomics Reveals Novel Biomarkers for Early Diagnosis of Hepatocellular Carcinoma.

Hepatocellular carcinoma (HCC) is a highly prevalent malignant tumor in China, and early diagnosis critically affects the prognosis. Current imaging and pathological biopsy techniques have limitations, including high invasiveness and limited accessibility, while the insufficient sensitivity of serum biomarkers (such as AFP) restricts their use in early screening. In this study, using the Olink proteomics platform based on the proximity extension assay (PEA), we screened for hepatocellular carcinoma-related differentially expressed proteins (DEPs) and constructed a multiprotein diagnostic model. In the discovery cohort, we included 15 patients with newly diagnosed HCCs and 16 healthy controls. DEPs were identified using Olink, and their diagnostic performance was analyzed to identify the candidate biomarkers. In an independent validation cohort, including 116 HCC patients (50 early stage, 66 late stage) and 83 healthy controls, we further validated the expression levels and diagnostic performance of identified proteins─C1QA and GFER. The C1QA and GFER expression levels were significantly higher in the serum of patients with early and late HCC stages compared to healthy controls. By constructing a multiprotein diagnostic model, we identified C1QA, GFER, and AFP as the optimal diagnostic combination, demonstrating a combined diagnostic AUC of 0.92 and 0.99 for early-stage and advanced-stage HCC, respectively.

Humans↗

PubMed2026

Large-Scale Plasma Proteomics Reveals Preclinical Biomarkers of Incident Severe Liver Disease.

The absence of robust biomarkers for early detection of severe liver disease (SLD) highlights the critical need for high-throughput proteomics-driven discovery. In this prospective cohort study, we aimed to identify plasma protein signatures associated with incident SLD and assess their clinical utility. Using the large-scale Olink Explore 1536 platform, we quantified 1461 plasma proteins in 46951 participants from the UK Biobank community-based cohort without baseline liver disease. Over a median follow-up of 14.1 years, we identified 490 proteins significantly associated with incident SLD risk. Growth differentiation factor 15 (GDF15) emerged as the strongest predictor, achieving a C-index of 0.80 and outperforming conventional clinical indices (LiverRisk score: 0.75; FIB-4: 0.68; APRI: 0.68). Temporal trajectories revealed that GDF15 levels began increasing up to 10 years before diagnosis, with progressive elevation as the diagnosis timepoint approached. Mendelian randomization analysis supported genetic associations linking higher protein levels of GDF15, FABP1, SPON2, CHI3L1, and PIGR with SLD risk. Our large-scale proteome-wide study not only reveals significant proteomic changes preceding SLD diagnosis but also establishes GDF15 as both a promising preclinical biomarker, opening new avenues for early intervention in at-risk individuals.

Humans↗

PubMed2026

Cardiac remodelling and dysfunction in cancer patients receiving cardiotoxic therapies: proteomic and metabolomic profiling.

BACKGROUND AND AIMS: The objective of this study was to define the relationships between the circulating proteome and metabolome with cardiac structure and function in patients with breast cancer receiving cardiotoxic therapies. METHODS: Proteomics and metabolomics profiling was performed in a longitudinal, prospective cohort study of breast cancer patients receiving anthracyclines and/or trastuzumab, using the Olink Explore 3072 platform and rapid liquid chromatography-mass spectrometry, respectively. Multivariable linear mixed-effect models evaluated the contemporaneous (same visit) and lagged (subsequent visit) associations between repeated measures of individual proteins or metabolites with quantitative echocardiographic measures of cardiac structure [left ventricular (LV) mass and left atrial volume index] and function [LV ejection fraction (LVEF), longitudinal and circumferential strain, E/e', and ventricular-arterial coupling]. Cox regression and pathway enrichment analyses were conducted for biomarkers demonstrating significant associations with cardiac function. RESULTS: Across 547 breast cancer participants (median age 50 years), 203 unique proteins and 16 unique metabolites were significantly associated with measures of cardiac structure and function in contemporaneous and lagged analyses. Notably, cathepsin C was associated with LVEF [false discovery rate (FDR), P = .017], longitudinal strain (FDR, P = .046), left atrial volume index (FDR, P = .035), and incident cardiac dysfunction, defined by an LVEF decline ≥10% to <50% (hazard ratio .61, 95% confidence interval .41, .90). The 147 proteins associated with cardiac function were enriched in biological processes reflective of protein deubiquitination, protein modification by small protein removal, macromolecule catabolic processes, and global metabolic pathways. Individual metabolites significantly associated with cardiac function (LVEF, longitudinal strain) included n-acetylglutamine, aspartic acid, acetylasparagine, alanyl-alanine, and prolyl-glycine (FDR, P-value < .001), and belonged to amino acids and derivatives and peptides. CONCLUSIONS: These findings provide translational insights into cancer therapy-related cardiac dysfunction and remodelling and identify potential new biomarkers of cardiotoxicity. There is an important need for validation of these findings and a deeper understanding of the biology of these biomarkers.

Humans↗

PubMed2026

Esketamine multi-omic biomarker evaluation in major depressive disorder (EMBER-MDD): concept, objectives and methodologies of a non-clinical investigator-initiated study.

Treatment resistance (TR) in major depressive disorder (MDD) affects a substantial minority of patients and is hard to recognize early, delaying intensified care. The Esketamine multi-omic biomarker evaluation in MDD (EMBER-MDD) is a non-interventional, investigator-initiated, in-vitro study within the EU Psych-STRATA programme, analyzing biospecimens collected in the randomized INTENSIFY study and the mirror OBS-TR cohort after participants complete treatment. EMBER-MDD aims to discover individual-omic and integrated multi-omic (hypothesis-free) biomarkers and signatures associated with TR risk, and molecular correlates of clinical response to esketamine nasal spray versus treatment as usual (TAU). Biomaterials will derive from approximately 420 adults with MDD (estimated n = 210 esketamine; n = 210 TAU) and include whole blood, RNA-stabilized whole blood, plasma and serum, sampled at baseline and, when feasible, during and after treatment (up to ~ 5,040 aliquots stored at - 80 °C). Genomics will use baseline DNA genotyping on Illumina Infinium GSA v3.0+MD arrays; epigenomics will profile genome-wide DNA methylation across time points using MethylationEPIC v2.0; transcriptomics will employ mRNA-seq (NovaSeq X/ X Plus); and proteomics/ metabolomics will be generated using high-throughput Olink and/ or Biocrates platforms. Each layer will undergo state-of-the-art preprocessing and analyses (e.g., GWAS/ PRS, EWAS, differential expression, WGCNA, pathway and network analyses), followed by integrative strategies including QTL mapping (meQTL/ eQTL/ pQTL/ mQTL) and intermediate-fusion machine learning with nested cross-validation, explainable AI (SHAP/ LIME) and treatment-effect modelling. All outputs are research-only and will not support individual efficacy, tolerability, or clinical decision-making. The study will deliver robust biosignatures and mechanistic hypotheses to guide future validation and inform stratified, molecularly guided intervention strategies in subsequent prospective trials. Trial registration number: 2023-506617-21-00 and 2025-178-f-S.

Humans↗

PubMed2026

DNA Methylation and Proteomic Profiling of Postmortem Brain Tissue Reveals Epigenetic Dysregulation and Neuroinflammatory in Fragile X-associated Tremor/Ataxia Syndrome (FXTAS).

BACKGROUND: Fragile X-associated Tremor/Ataxia Syndrome (FXTAS) is a late-onset neurodegenerative disorder caused by FMR1 premutation CGG repeat expansions (55-200 repeats). The epigenetic landscape of the FXTAS brain remains uncharacterized. We performed genome-wide DNA methylation profiling of postmortem prefrontal cortex tissue to identify differentially methylated positions (DMPs) and candidate genes, and sought protein-level support for a neuroinflammatory signal. METHODS: DNA methylation was profiled in postmortem prefrontal cortex (Brodmann area 9) from 27 male FXTAS cases and 29 male controls using the Illumina MethylationEPIC array (EPICv1 and EPICv2 platforms), merging 721,802 common probes. Surrogate variable analysis (SVA) controlled for confounders. DMPs were defined by |Δβ| > 0.10 and FDR < 0.05; exploratory Reactome 2024 pathway analysis was performed on the DMP-associated gene list. Targeted proteomic profiling was performed in the same brain region using the Olink (proximity extension assay) Inflammation panel in 9 FXTAS cases and 12 controls, with SVA-adjusted differential abundance analysis, and concordance assessment against a prior mass spectrometry dataset. RESULTS: We identified 108 significant cg-type DMPs mapping to 80 genes (50 hypermethylated, 58 hypomethylated in FXTAS). The strongest signal was CYP2E1 (7 concordant hypomethylated DMPs, mean Δβ = -0.143), an oxidative stress gene also implicated in Parkinson's disease. FTCD, a one-carbon cycle enzyme, carried 5 hypermethylated DMPs (mean Δβ = +0.210). A cluster of DMP-associated genes with established roles in innate immune and NF-κB signaling, TRAF3 (the single most significant DMP among the inflammation genes, hypermethylated), BATF, RCOR1, and MSI2; they pointed toward neuroinflammatory dysregulation. Additional genes included LINGO1 (myelination inhibitor), SYT3 (synaptic vesicle), and SLC39A4 (zinc transporter). Exploratory Reactome enrichment using the DMP-associated gene set nominated themes including neuroinflammation resolution, axonal growth inhibition, zinc homeostasis, and CYP2E1 metabolism at nominal significance (p<0.05); however, the gene-to-pathway mapping rate was low and no pathway survived correction for multiple testing. Olink proteomic analysis independently identified 60 significantly altered inflammation proteins (59 downregulated), including CXCL8, CXCL10, IL6, IL15, IL18, TLR3, IRAK1/4, and complement C1QA, which were directionally concordant with prior mass spectrometry data. CONCLUSIONS: This integrated study reveals a genome-wide epigenetic signature in the FXTAS prefrontal cortex implicating oxidative stress, myelination failure, zinc dysregulation, one-carbon cycle disruption, and most notably a coordinated set of epigenetically altered genes governing innate immune and NF-κB signaling. Convergence of TRAF3 hypermethylation with independent downregulation of TLR3 and NF-κB-pathway proteins at the protein level supports a coherent, cross-platform model of dysregulated neuroinflammatory signaling in FXTAS, identified here through individual gene- and protein-level convergence rather than formal pathway enrichment. FTCD hypermethylation proposes a self-reinforcing epigenetic loop via SAM depletion. These multi-omic findings establish FXTAS as a disorder of pervasive epigenetic reprogramming and nominate candidate genes for future mechanistic and therapeutic investigation.

CYP2E1↗

PubMed2024

Secretome Analysis Using Affinity Proteomics and Immunoassays: A Focus on Tumor Biology.

The study of the cellular secretome using proteomic techniques continues to capture the attention of the research community across a broad range of topics in biomedical research. Due to their untargeted nature, independence from the model system used, historically superior depth of analysis, as well as comparative affordability, mass spectrometry-based approaches traditionally dominate such analyses. More recently, however, affinity-based proteomic assays have massively gained in analytical depth, which together with their high sensitivity, dynamic range coverage as well as high throughput capabilities render them exquisitely suited to secretome analysis. In this review, we revisit the analytical challenges implied by secretomics and provide an overview of affinity-based proteomic platforms currently available for such analyses, using the study of the tumor secretome as an example for basic and translational research.

Humans↗

PubMed2026

Machine learning-based clinical prediction model and multi-omics integration for assessing pancreatic cancer risk in new-onset diabetes.

BACKGROUND: Given that pancreatic cancer (PC) is typically diagnosed at an advanced stage but is often preceded by new-onset diabetes mellitus (NODM), providing a window for early detection, we sought to develop and validate an interpretable machine-learning model integrated with multi-omics profiling to identify early biomarkers of NODM-associated PC. METHODS: In a population-based cohort, individuals with NODM-associated PC and NODM without PC were identified and randomly divided (70:30) into training and validation sets after feature selection. Eight machine learning (ML) classifiers were compared using fivefold cross-validation, and model performance was evaluated in terms of discrimination, calibration, and decision curve–based clinical utility. We evaluated interpretability using the Shapley additive explanations (SHAP) analyses. Mechanistically, Olink proteomic profiling and metabolomics were analyzed through clinical classifications and model-defined risk strata. RESULTS: Categorical boosting achieved the best performance in the independent validation set (AUROC = 0.844). The NODM cohort was stratified into high- (n = 2,362) and low-risk (n = 5,030) groups, and internal validation together with SHAP analyses demonstrated consistent model performance and identified clinically interpretable predictors. Proteomic and metabolomic analyses under clinical and risk-based grouping identified 39 overlapping differentially expressed proteins and 145 overlapping metabolites with enriched across 11 shared KEGG pathways. Cross-platform validation highlighted PLTP, CRTAC1, and ITGAV as serum biomarkers with a strong potential for early NODM-PC detection. CONCLUSIONS: We developed an interpretable ML framework centered on NODM enables practical risk stratification for early PC detection by multi-omics and provides a pathway of ML-based triage followed by biomarker confirmation for earlier detection and diagnosis.

Humans↗

PubMed2026

Proteomics identify disease-associated variants in patients with rare diseases undiagnosed after genome sequencing.

Despite the introduction of genome sequencing (GS) for rare disease diagnostics, a genetic cause is not identified in most patients. Here, we explored the potential of proteomics to improve the diagnostic yield in 424 patients with rare diseases from the 100,000 Genomes Project (100kGP) without a genetic diagnosis. Serum proteomic profiling was performed using the Olink Explore 1536 assay (N = 1463 proteins). For 13 patients without genetic diagnoses, detection of lower serum protein "outliers" (z-score < -2) led to confirmed genetic diagnoses by resolving variants of uncertain significance or prioritizing genes for targeted GS reanalysis. For 23 additional patients without genetic diagnoses (64% of findings), we identified candidate gene-disease links and variants through convergent evidence from lower protein outliers and variants ranked through the variant prioritization tool Exomiser. For example, we identified a candidate heterozygous missense variant [Genome Aggregation Database (gnomAD) minor allele frequency = 0.006%] in tyrosine kinase with immunoglobulin-like and epidermal growth factor homology domains 1 (TIE1) that was only present in a patient with lower TIE1 serum abundance (z-score = -5.12) and their father, both of whom were affected by the same monogenic cardiac disorder, but in no other individuals from the 100kGP. Missense (52.5%) and splice region (27.5%) variants accounted for most diagnostic or candidate variants prioritized. This proof-of-principle study demonstrated that serum proteomics can support rare disease diagnosis and identify disease-causing genes in patients undiagnosed after GS, although successful implementation will likely depend on tissue specificity of protein expression, detectability in blood, proteomic platform coverage, and sensitivity.

Humans↗