Plasma proteins capture dynamic physiological processes and may offer more immediate insight into disease risk than static genetic predictors. We evaluated the predictive utility of proteomic risk scores (ProRS) versus polygenic risk scores (PRS) across 301 phenotypes in 39,843 participants from the UK Biobank Pharma Proteomics Project. ProRS, trained on prevalent cases, were tested for incident disease and benchmarked against PRS derived from genome-wide association statistics. Among 268 phenotypes with informative signals, ProRS outperformed PRS in 88% of traits (a median C-index improvement of 9.6%), showing strongest gains for circulatory, metabolic, and immune conditions. Combined models further improved prediction, particularly for traits with higher heritability. Longitudinal analyses showed that ProRS values were elevated years before diagnosis. External validation in 841 Penn Medicine BioBank participants confirmed consistent performance and transferability, with AUC improvements up to 4.18% over PRS alone. Plasma proteomic profiling provides complementary, temporally responsive information that enhances individual-level disease prediction.

INTRODUCTION: Alzheimer's disease (AD) is a complex neurodegenerative disorder with heterogeneous genetic and molecular underpinnings. Polygenic scores (PGS) capture little of this complexity. METHODS: We conducted genome‐, transcriptome‐, and proteome‐wide association studies (G/T/PWAS) on 15,480 individuals from the Alzheimer's Disease Sequencing Project R4 (ADSP) to identify AD‐associated signals, followed by pathway enrichment analysis. Integrative risk models (IRMs) were developed using genetically regulated components of gene and protein expression and clinical covariates. Elastic‐net logistic regression and random forest classifiers were evaluated using standard metrics and compared against baseline PGS. RESULTS: Known and novel signals were identified via G/T/PWAS. Enrichment analyses highlighted cholesterol and immune signaling pathways. The best‐performing IRM, random forest with transcriptomic and covariate features, achieved area under the receiver operating characteristic (AUROC) of 0.703 and area under the precision‐recall curve (AUPRC) of 0.622, significantly outperforming PGS and baseline models. DISCUSSION: Integrating univariate discovery approaches with multivariate modeling enhances AD risk prediction and offers novel insights into underlying biological processes. Highlights: Identified novel contributions to Alzheimer's disease (AD) from a multi‐omics perspective. Integrated genome‐wide association studies (GWAS), transcriptome‐wide association studies (TWAS), and proteome‐wide association studies (PWAS) in a unified association study framework. Developed a method for predicting heritable risk of late‐onset AD. Demonstrated that ancestry‐aware modeling improves AD risk prediction accuracy.

Coronary microvascular disease (CMVD) is an underdiagnosed but significant contributor to the burden of ischemic heart disease, characterized by angina and myocardial infarction. The development of risk prediction models such as polygenic risk scores (PRS) for CMVD has been limited by a lack of large-scale genome-wide association studies (GWAS). However, there is significant overlap between CMVD and enrollment criteria for coronary artery disease (CAD) GWAS. In this study, we developed CMVD PRS models by selecting variants identified in a CMVD GWAS and applying weights from an external CAD GWAS, using CMVD-associated loci as proxies for the genetic risk. We integrated plasma proteomics, clinical measures from perfusion PET imaging, and PRS to evaluate their contributions to CMVD risk prediction in comprehensive machine and deep learning models. We then developed a novel unsupervised endotyping framework for CMVD from perfusion PET-derived myocardial blood flow data, revealing distinct patient subgroups beyond traditional case-control definitions. This imaging-based stratification substantially improved classification performance alongside plasma proteomics and PRS, achieving AUROCs between 0.65 and 0.73 per class, significantly outperforming binary classifiers and existing clinical models, highlighting the potential of this stratification approach to enable more precise and personalized diagnosis by capturing the underlying heterogeneity of CMVD. This work represents the first application of imaging-based endotyping and the integration of genetic and proteomic data for CMVD risk prediction, establishing a framework for multimodal modeling in complex diseases.

Metabolic dysfunction-associated Fatty Liver Disease (MAFLD) has emerged as one of the leading cardiometabolic diseases. Friend of GATA2 (FOG2) is a transcriptional co-regulator that has been shown to regulate hepatic lipid metabolism and accumulation. Using meta-analysis from several different biobank datasets, we identified a coding variant of FOG2 (rs28374544, A1969G, S657G) predominantly found in individuals of African ancestry (minor allele frequency~20%), which is associated with liver failure/cirrhosis phenotype and liver injury. To gain insight into potential pathways associated with this variant, we interrogated a previously published genomics dataset of 38 human induced pluripotent stem cell (iPSCs) lines differentiated into hepatocytes (iHeps). Using Differential Gene Expression Analysis and Gene Set Enrichment Analysis, we identified the mTORC1 pathway as differentially regulated between iHeps from individuals with and without the variant. Transient lipid-based transfections were performed on the human hepatoma cell line (Huh7) using wild-type FOG2 and FOG2^S657G and demonstrated that FOG2S^657G increased mTORC1 signaling, de novo lipogenesis, and cellular triglyceride synthesis and mass. In addition, we observed a significant downregulation of oxidative phosphorylation in FOG2^S657G cells in fatty acid-loaded cells but not untreated cells, suggesting that FOG2^S657G may also reduce fatty acid to promote lipid accumulation. Taken together, our multi-pronged approach suggests a model whereby the FOG2^S657G may promote MAFLD through mTORC1 activation, increased de novo lipogenesis, and lipid accumulation. Our results provide insights into the molecular mechanisms by which FOG2^S657G may affect the complex molecular landscape underlying MAFLD.

Alzheimer’s Disease (AD) is the most prevalent condition that impacts the aging population, with no effective treatment or singular underlying causal factor identified. As a complex disease, characterizing the genetic risk of developing AD has proven to be difficult; polygenic scores (PGS) exclusively use common variants which fail to fully capture disease heterogeneity. This study used univariate and multivariate approaches to characterize AD risk. Genome-, transcriptome-, and proteome-wide association studies (GWAS, TWAS, and PWAS) were conducted on 15,480 individuals from the Alzheimer’s Disease Sequencing Project (ADSP) R4 release to identify AD-associated signals, followed by pathway enrichment analysis. Integrative risk models (IRMs) were developed using genetically-regulated components of gene and protein expression and clinical covariates. Elastic-net logistic regression and random forest classifiers were evaluated using area under the receiver operating characteristic (AUROC), area under the precision-recall curve (AUPRC), F1-score, and balanced accuracy. These IRMs were compared against baseline PGS and covariate models. We identified 104 genomic, 319 transcriptomic, and 17 proteomic associations with AD under significant thresholds. Putatively novel associations were enriched in signaling, myeloid differentiation, and immune pathways. The best-performing IRM, random forest with transcriptomic and covariate features, achieved an AUROC of 0.703 and AUPRC of 0.622, significantly outperforming PGS and baseline models. Integrating univariate discovery approaches with multivariate modeling enhances AD risk prediction and offers insights into underlying biological processes.

Multi-omics data are instrumental in obtaining a comprehensive picture of complex biological systems. This is particularly useful for women’s health conditions, such as endometriosis which has been historically understudied despite having a high prevalence (around 10% of women of reproductive age). Subsequently, endometriosis has limited genetic characterization: current genome-wide association studies explain only 11% of its 47% total estimated heritability. Graph representations provide an intuitive and meaningful way to relate concepts across diverse data sources and address fundamental sparsity and dimensionality challenges with multi-omics data analysis. Here we present DRIVE-KG (Disease Risk Inference and Variant Exploration-Knowledge Graph), which uses a heterogeneous graph representation to integrate biological data from multi-omics datasets: dbSNP, NCBI Human Gene, Omics Pred, GTEx, and Open Targets. We drew directly from the knowledge captured in these data, using nodes to represent genes, single nucleotide polymorphisms, proteins, and phenotypes, and edges to represent relationships between these concepts. We trained two models using DRIVE-KG: a link prediction model to suggest associations between SNPs and two pilot phenotypes (endometriosis and obesity), and a graph convolutional network (GCN) to classify patient-level endometriosis status. We conducted the patient-level classification using data from 1,441 Penn Medicine BioBank participants with gold standard chart-reviewed endometriosis status. The link prediction model uncovered 66 high-confidence (score ≥ 0.95) previously unreported SNP-endometriosis associations. Many of these variants were linked to obesity/body mass index traits (24.2%), lipid metabolism (6%), and depressive disorders (4.5%), showing agreement with emerging hypotheses about endometriosis etiology. In contrast, 11% of the 149 high confidence, candidate SNP-obesity associations (score ≥ 0.9888) were in LD with known obesity associations. The GCN to classify patient endometriosis status had an AUPRC of 0.738 compared to 0.679 for a genetic risk score. Despite this moderate improvement, we found that the GCN learned meaningful stratification of underlying adenomyosis signal and severe grades of endometriosis. We have demonstrated that heterogeneous integration of multi-omics data is valuable for diverse downstream tasks—including discovery and clinical prediction—particularly for understudied diseases where traditional genomic approaches are insufficient.