The association of cardiometabolic, diet and lifestyle parameters with plasma glucagon-like peptide-1: An IMI DIRECT study.

CONTEXT: The role of glucagon-like peptide-1(GLP-1) in Type 2 diabetes (T2D) and obesity is not fully understood. OBJECTIVE: We investigate the association of cardiometabolic, diet and lifestyle parameters on fasting and postprandial GLP-1 in people at risk of, or living with, T2D. METHOD: We analysed cross-sectional data from the two Innovative Medicines Initiative (IMI) Diabetes Research on Patient Stratification (DIRECT) cohorts, cohort 1(n=2127) individuals at risk of diabetes; cohort 2 (n=789) individuals with new-onset of T2D. RESULTS: Our multiple regression analysis reveals that fasting total GLP-1 is associated with an insulin resistant phenotype and observe a strong independent relationship with male sex, increased adiposity and liver fat particularly in the prediabetes population. In contrast, we showed that incremental GLP-1 decreases with worsening glycaemia, higher adiposity, liver fat, male sex and reduced insulin sensitivity in the prediabetes cohort. Higher fasting total GLP-1 was associated with a low intake of wholegrain, fruit and vegetables inpeople with prediabetes, and with a high intake of red meat and alcohol in people with diabetes. CONCLUSION: These studies provide novel insights into the association between fasting and incremental GLP-1, metabolic traits of diabetes and obesity, and dietary intake and raise intriguing questions regarding the relevance of fasting GLP-1 in the pathophysiology T2D.

Proteomic analysis of 92 circulating proteins and their effects in cardiometabolic diseases.

BACKGROUND: Human plasma contains a wide variety of circulating proteins. These proteins can be important clinical biomarkers in disease and also possible drug targets. Large scale genomics studies of circulating proteins can identify genetic variants that lead to relative protein abundance. METHODS: We conducted a meta-analysis on genome-wide association studies of autosomal chromosomes in 22,997 individuals of primarily European ancestry across 12 cohorts to identify protein quantitative trait loci (pQTL) for 92 cardiometabolic associated plasma proteins. RESULTS: We identified 503 (337 cis and 166 trans) conditionally independent pQTLs, including several novel variants not reported in the literature. We conducted a sex-stratified analysis and found that 118 (23.5%) of pQTLs demonstrated heterogeneity between sexes. The direction of effect was preserved but there were differences in effect size and significance. Additionally, we annotate trans-pQTLs with nearest genes and report plausible biological relationships. Using Mendelian randomization, we identified causal associations for 18 proteins across 19 phenotypes, of which 10 have additional genetic colocalization evidence. We highlight proteins associated with a constellation of cardiometabolic traits including angiopoietin-related protein 7 (ANGPTL7) and Semaphorin 3F (SEMA3F). CONCLUSION: Through large-scale analysis of protein quantitative trait loci, we provide a comprehensive overview of common variants associated with plasma proteins. We highlight possible biological relationships which may serve as a basis for further investigation into possible causal roles in cardiometabolic diseases.

Discovery of drug-omics associations in type 2 diabetes with generative deep-learning models.

The application of multiple omics technologies in biomedical cohorts has the potential to reveal patient-level disease characteristics and individualized response to treatment. However, the scale and heterogeneous nature of multi-modal data makes integration and inference a non-trivial task. We developed a deep-learning-based framework, multi-omics variational autoencoders (MOVE), to integrate such data and applied it to a cohort of 789 people with newly diagnosed type 2 diabetes with deep multi-omics phenotyping from the DIRECT consortium. Using in silico perturbations, we identified drug-omics associations across the multi-modal datasets for the 20 most prevalent drugs given to people with type 2 diabetes with substantially higher sensitivity than univariate statistical tests. From these, we among others, identified novel associations between metformin and the gut microbiota as well as opposite molecular responses for the two statins, simvastatin and atorvastatin. We used the associations to quantify drug-drug similarities, assess the degree of polypharmacy and conclude that drug effects are distributed across the multi-omics modalities.

Multiallelic Copy Number Variation in ORM1 is Associated with Plasma Cell-Free DNA Levels as an Intermediate Phenotype for Venous Thromboembolism.

Venous thromboembolism (VTE) is a common disease with high heritability. However, only a small portion of the genetic variance of VTE can be explained by known genetic risk factors. Neutrophil extracellular traps (NETs) have been associated with prothrombotic activity. Therefore, the genetic basis of NETs could reveal novel risk factors for VTE. A recent genome-wide association study of plasma cell-free DNA (cfDNA) levels in the Genetic Analysis of Idiopathic Thrombophilia 2 (GAIT-2) Project showed a significant associated locus near ORM1. We aimed to further explore this candidate region by next-generation sequencing, copy number variation (CNV) quantification, and expression analysis using an extreme phenotype sampling design involving 80 individuals from the GAIT-2 Project. The RETROVE study with 400 VTE cases and 400 controls was used to replicate the results. A total of 105 genetic variants and a multiallelic CNV (mCNV) spanning ORM1 were identified in GAIT-2. Of these, 17 independent common variants, a region of 22 rare variants, and the mCNV were significantly associated with cfDNA levels. In addition, eight of these common variants and the mCNV influenced ORM1 expression. The association of the mCNV and cfDNA levels was replicated in RETROVE (p-value = 1.19 × 10-6). Additional associations between the mCNV and thrombin generation parameters were identified. Our results reveal that increased mCNV dosages in ORM1 decreased gene expression and upregulated cfDNA levels. Therefore, the mCNV in ORM1 appears to be a novel marker for cfDNA levels, which could contribute to VTE risk.

Genetic Landscape of the ACE2 Coronavirus Receptor.

BACKGROUND: SARS-CoV-2, the causal agent of COVID-19, enters human cells using the ACE2 (angiotensin-converting enzyme 2) protein as a receptor. ACE2 is thus key to the infection and treatment of the coronavirus. ACE2 is highly expressed in the heart and respiratory and gastrointestinal tracts, playing important regulatory roles in the cardiovascular and other biological systems. However, the genetic basis of the ACE2 protein levels is not well understood. METHODS: We have conducted the largest genome-wide association meta-analysis of plasma ACE2 levels in >28 000 individuals of the SCALLOP Consortium (Systematic and Combined Analysis of Olink Proteins). We summarize the cross-sectional epidemiological correlates of circulating ACE2. Using the summary statistics-based high-definition likelihood method, we estimate relevant genetic correlations with cardiometabolic phenotypes, COVID-19, and other human complex traits and diseases. We perform causal inference of soluble ACE2 on vascular disease outcomes and COVID-19 severity using mendelian randomization. We also perform in silico functional analysis by integrating with other types of omics data. RESULTS: We identified 10 loci, including 8 novel, capturing 30% of the heritability of the protein. We detected that plasma ACE2 was genetically correlated with vascular diseases, severe COVID-19, and a wide range of human complex diseases and medications. An X-chromosome cis-protein quantitative trait loci-based mendelian randomization analysis suggested a causal effect of elevated ACE2 levels on COVID-19 severity (odds ratio, 1.63 [95% CI, 1.10-2.42]; P=0.01), hospitalization (odds ratio, 1.52 [95% CI, 1.05-2.21]; P=0.03), and infection (odds ratio, 1.60 [95% CI, 1.08-2.37]; P=0.02). Tissue- and cell type-specific transcriptomic and epigenomic analysis revealed that the ACE2 regulatory variants were enriched for DNA methylation sites in blood immune cells. CONCLUSIONS: Human plasma ACE2 shares a genetic basis with cardiovascular disease, COVID-19, and other related diseases. The genetic architecture of the ACE2 protein is mapped, providing a useful resource for further biological and clinical studies on this coronavirus receptor.

Integrated GWAS and Gene Expression Suggest ORM1 as a Potential Regulator of Plasma Levels of Cell-Free DNA and Thrombosis Risk.

Plasma cell-free DNA (cfDNA) is a surrogate marker of neutrophil extracellular traps (NETs) that contribute to immunothrombosis. There is growing interest about the mechanisms underlying NET formation and elevated cfDNA, but little is known about the factors involved. We aimed to identify genes involved in the regulation of cfDNA levels using data from the Genetic Analysis of Idiopathic Thrombophilia (GAIT-2) Project.Imputed genotypes, whole blood RNA-Seq data, and plasma cfDNA quantification were available for 935 of the GAIT-2 participants from 35 families with idiopathic thrombophilia. We performed heritability and GWAS analysis for cfDNA. The heritability of cfDNA was 0.26 (p = 3.7 × 10-6), while the GWAS identified a significant association (rs1687391, p = 3.55 × 10-10) near the ORM1 gene, on chromosome 9. An eQTL (expression quantitative trait loci) analysis revealed a significant association between the lead GWAS variant and the expression of ORM1 in whole blood (p = 6.14 × 10-9). Additionally, ORM1 expression correlated with levels of cfDNA (p = 4.38 × 10-4). Finally, genetic correlation analysis between cfDNA and thrombosis identified a suggestive association (ρ g = 0.43, p = 0.089).All in all, we show evidence of the role of ORM1 in regulating cfDNA levels in plasma, which might contribute to the susceptibility to thrombosis through mechanisms of immunothrombosis.

Four groups of type 2 diabetes contribute to the etiological and clinical heterogeneity in newly diagnosed individuals: An IMI DIRECT study.

The presentation and underlying pathophysiology of type 2 diabetes (T2D) is complex and heterogeneous. Recent studies attempted to stratify T2D into distinct subgroups using data-driven approaches, but their clinical utility may be limited if categorical representations of complex phenotypes are suboptimal. We apply a soft-clustering (archetype) method to characterize newly diagnosed T2D based on 32 clinical variables. We assign quantitative clustering scores for individuals and investigate the associations with glycemic deterioration, genetic risk scores, circulating omics biomarkers, and phenotypic stability over 36 months. Four archetype profiles represent dysfunction patterns across combinations of T2D etiological processes and correlate with multiple circulating biomarkers. One archetype associated with obesity, insulin resistance, dyslipidemia, and impaired ß cell glucose sensitivity corresponds with the fastest disease progression and highest demand for anti-diabetic treatment. We demonstrate that clinical heterogeneity in T2D can be mapped to heterogeneity in individual etiological processes, providing a potential route to personalized treatments.

Deletion of ABCB10 in beta-cells protects from high-fat diet induced insulin resistance.

OBJECTIVE: The contribution of beta-cell dysfunction to type 2 diabetes (T2D) is not restricted to insulinopenia in the late stages of the disease. Elevated fasting insulinemia in normoglycemic humans is a major factor predicting the onset of insulin resistance and T2D, demonstrating an early alteration of beta-cell function in T2D. Moreover, an early and chronic increase in fasting insulinemia contributes to insulin resistance in high-fat diet (HFD)-fed mice. However, whether there are genetic factors that promote beta-cell-initiated insulin resistance remains undefined. Human variants of the mitochondrial transporter ABCB10, which regulates redox by increasing bilirubin synthesis, have been associated with an elevated risk of T2D. The effects of T2D ABCB10 variants on ABCB10 expression and the actions of ABCB10 in beta-cells are unknown. METHODS: The expression of beta-cell ABCB10 was analyzed in published transcriptome datasets from human beta-cells carrying the T2D-risk ABCB10 variant. Insulin sensitivity, beta-cell proliferation, and secretory function were measured in beta-cell-specific ABCB10 KO mice (Ins1Cre-Abcb10flox/flox). The short-term role of beta-cell ABCB10 activity on glucose-stimulated insulin secretion (GSIS) was determined in isolated islets. RESULTS: Carrying the T2Drisk allele G of ABCB10 rs348330 variant was associated with increased ABCB10 expression in human beta-cells. Constitutive deletion of Abcb10 in beta-cells protected mice from hyperinsulinemia and insulin resistance by limiting HFD-induced beta-cell expansion. An early limitation in GSIS and H2O2-mediated signaling caused by elevated ABCB10 activity can initiate an over-compensatory expansion of beta-cell mass in response to HFD. Accordingly, increasing ABCB10 expression was sufficient to limit GSIS capacity. In health, ABCB10 protein was decreased during islet maturation, with maturation restricting beta-cell proliferation and elevating GSIS. Finally, ex-vivo and short-term deletion of ABCB10 in islets isolated from HFD-fed mice increased H2O2 and GSIS, which was reversed by bilirubin treatments. CONCLUSIONS: Beta-cell ABCB10 is required for HFD to induce insulin resistance in mice by amplifying beta-cell mass expansion to maladaptive levels that cause fasting hyperinsulinemia.

Profiles of Glucose Metabolism in Different Prediabetes Phenotypes, Classified by Fasting Glycemia, 2-Hour OGTT, Glycated Hemoglobin, and 1-Hour OGTT: An IMI DIRECT Study.

Differences in glucose metabolism among categories of prediabetes have not been systematically investigated. In this longitudinal study, participants (N = 2,111) underwent a 2-h 75-g oral glucose tolerance test (OGTT) at baseline and 48 months. HbA1c was also measured. We classified participants as having isolated prediabetes defect (impaired fasting glucose [IFG], impaired glucose tolerance [IGT], or HbA1c indicative of prediabetes [IA1c]), two defects (IFG+IGT, IFG+IA1c, or IGT+IA1c), or all defects (IFG+IGT+IA1c). ß-Cell function (BCF) and insulin sensitivity were assessed from OGTT. At baseline, in pooling of participants with isolated defects, they showed impairment in both BCF and insulin sensitivity compared with healthy control subjects. Pooled groups with two or three defects showed progressive further deterioration. Among groups with isolated defect, those with IGT showed lower insulin sensitivity, insulin secretion at reference glucose (ISRr), and insulin secretion potentiation (P < 0.002). Conversely, those with IA1c showed higher insulin sensitivity and ISRr (P < 0.0001). Among groups with two defects, we similarly found differences in both BCF and insulin sensitivity. At 48 months, we found higher type 2 diabetes incidence for progressively increasing number of prediabetes defects (odds ratio >2, P < 0.008). In conclusion, the prediabetes groups showed differences in type/degree of glucometabolic impairment. Compared with the pooled group with isolated defects, those with double or triple defect showed progressive differences in diabetes incidence.

Genome-Wide Association Analysis of Pancreatic Beta-Cell Glucose Sensitivity.

CONTEXT: Pancreatic beta-cell glucose sensitivity is the slope of the plasma glucose-insulin secretion relationship and is a key predictor of deteriorating glucose tolerance and development of type 2 diabetes. However, there are no large-scale studies looking at the genetic determinants of beta-cell glucose sensitivity. OBJECTIVE: To understand the genetic determinants of pancreatic beta-cell glucose sensitivity using genome-wide meta-analysis and candidate gene studies. DESIGN: We performed a genome-wide meta-analysis for beta-cell glucose sensitivity in subjects with type 2 diabetes and nondiabetic subjects from 6 independent cohorts (n = 5706). Beta-cell glucose sensitivity was calculated from mixed meal and oral glucose tolerance tests, and its associations between known glycemia-related single nucleotide polymorphisms (SNPs) and genome-wide association study (GWAS) SNPs were estimated using linear regression models. RESULTS: Beta-cell glucose sensitivity was moderately heritable (h2 ranged from 34% to 55%) using SNP and family-based analyses. GWAS meta-analysis identified multiple correlated SNPs in the CDKAL1 gene and GIPR-QPCTL gene loci that reached genome-wide significance, with SNP rs2238691 in GIPR-QPCTL (P value = 2.64 × 10-9) and rs9368219 in the CDKAL1 (P value = 3.15 × 10-9) showing the strongest association with beta-cell glucose sensitivity. These loci surpassed genome-wide significance when the GWAS meta-analysis was repeated after exclusion of the diabetic subjects. After correction for multiple testing, glycemia-associated SNPs in or near the HHEX and IGF2B2 loci were also associated with beta-cell glucose sensitivity. CONCLUSION: We show that, variation at the GIPR-QPCTL and CDKAL1 loci are key determinants of pancreatic beta-cell glucose sensitivity.

Processes Underlying Glycemic Deterioration in Type 2 Diabetes: An IMI DIRECT Study.

OBJECTIVE: We investigated the processes underlying glycemic deterioration in type 2 diabetes (T2D). RESEARCH DESIGN AND METHODS: A total of 732 recently diagnosed patients with T2D from the Innovative Medicines Initiative Diabetes Research on Patient Stratification (IMI DIRECT) study were extensively phenotyped over 3 years, including measures of insulin sensitivity (OGIS), ß-cell glucose sensitivity (GS), and insulin clearance (CLIm) from mixed meal tests, liver enzymes, lipid profiles, and baseline regional fat from MRI. The associations between the longitudinal metabolic patterns and HbA1c deterioration, adjusted for changes in BMI and in diabetes medications, were assessed via stepwise multivariable linear and logistic regression. RESULTS: Faster HbA1c progression was independently associated with faster deterioration of OGIS and GS and increasing CLIm; visceral or liver fat, HDL-cholesterol, and triglycerides had further independent, though weaker, roles (R 2 = 0.38). A subgroup of patients with a markedly higher progression rate (fast progressors) was clearly distinguishable considering these variables only (discrimination capacity from area under the receiver operating characteristic = 0.94). The proportion of fast progressors was reduced from 56% to 8-10% in subgroups in which only one trait among OGIS, GS, and CLIm was relatively stable (odds ratios 0.07-0.09). T2D polygenic risk score and baseline pancreatic fat, glucagon-like peptide 1, glucagon, diet, and physical activity did not show an independent role. CONCLUSIONS: Deteriorating insulin sensitivity and ß-cell function, increasing insulin clearance, high visceral or liver fat, and worsening of the lipid profile are the crucial factors mediating glycemic deterioration of patients with T2D in the initial phase of the disease. Stabilization of a single trait among insulin sensitivity, ß-cell function, and insulin clearance may be relevant to prevent progression.

Correction to: The role of physical activity in metabolic homeostasis before and after the onset of type 2 diabetes: an IMI DIRECT study.

Unfortunately, 'Present address' was omitted from one of the addresses provided for Mark I. McCarthy (#26).

Whole blood co-expression modules associate with metabolic traits and type 2 diabetes: an IMI-DIRECT study.

BACKGROUND: The rising prevalence of type 2 diabetes (T2D) poses a major global challenge. It remains unresolved to what extent transcriptomic signatures of metabolic dysregulation and T2D can be observed in easily accessible tissues such as blood. Additionally, large-scale human studies are required to further our understanding of the putative inflammatory component of insulin resistance and T2D. Here we used transcriptomics data from individuals with (n = 789) and without (n = 2127) T2D from the IMI-DIRECT cohorts to describe the co-expression structure of whole blood that mainly reflects processes and cell types of the immune system, and how it relates to metabolically relevant clinical traits and T2D. METHODS: Clusters of co-expressed genes were identified in the non-diabetic IMI-DIRECT cohort and evaluated with regard to stability, as well as preservation and rewiring in the cohort of individuals with T2D. We performed functional and immune cell signature enrichment analyses, and a genome-wide association study to describe the genetic regulation of the modules. Phenotypic and trans-omics associations of the transcriptomic modules were investigated across both IMI-DIRECT cohorts. RESULTS: We identified 55 whole blood co-expression modules, some of which clustered in larger super-modules. We identified a large number of associations between these transcriptomic modules and measures of insulin action and glucose tolerance. Some of the metabolically linked modules reflect neutrophil-lymphocyte ratio in blood while others are independent of white blood cell estimates, including a module of genes encoding neutrophil granule proteins with antibacterial properties for which the strongest associations with clinical traits and T2D status were observed. Through the integration of genetic and multi-omics data, we provide a holistic view of the regulation and molecular context of whole blood transcriptomic modules. We furthermore identified an overlap between genetic signals for T2D and co-expression modules involved in type II interferon signaling. CONCLUSIONS: Our results offer a large-scale map of whole blood transcriptomic modules in the context of metabolic disease and point to novel biological candidates for future studies related to T2D.

Post-load glucose subgroups and associated metabolic traits in individuals with type 2 diabetes: An IMI-DIRECT study.

AIM: Subclasses of different glycaemic disturbances could explain the variation in characteristics of individuals with type 2 diabetes (T2D). We aimed to examine the association between subgroups based on their glucose curves during a five-point mixed-meal tolerance test (MMT) and metabolic traits at baseline and glycaemic deterioration in individuals with T2D. METHODS: The study included 787 individuals with newly diagnosed T2D from the Diabetes Research on Patient Stratification (IMI-DIRECT) Study. Latent class trajectory analysis (LCTA) was used to identify distinct glucose curve subgroups during a five-point MMT. Using general linear models, these subgroups were associated with metabolic traits at baseline and after 18 months of follow up, adjusted for potential confounders. RESULTS: At baseline, we identified three glucose curve subgroups, labelled in order of increasing glucose peak levels as subgroup 1-3. Individuals in subgroup 2 and 3 were more likely to have higher levels of HbA1c, triglycerides and BMI at baseline, compared to those in subgroup 1. At 18 months (n = 651), the beta coefficients (95% CI) for change in HbA1c (mmol/mol) increased across subgroups with 0.37 (-0.18-1.92) for subgroup 2 and 1.88 (-0.08-3.85) for subgroup 3, relative to subgroup 1. The same trend was observed for change in levels of triglycerides and fasting glucose. CONCLUSIONS: Different glycaemic profiles with different metabolic traits and different degrees of subsequent glycaemic deterioration can be identified using data from a frequently sampled mixed-meal tolerance test in individuals with T2D. Subgroups with the highest peaks had greater metabolic risk.

Genetic variant effects on gene expression in human pancreatic islets and their implications for T2D.

Most signals detected by genome-wide association studies map to non-coding sequence and their tissue-specific effects influence transcriptional regulation. However, key tissues and cell-types required for functional inference are absent from large-scale resources. Here we explore the relationship between genetic variants influencing predisposition to type 2 diabetes (T2D) and related glycemic traits, and human pancreatic islet transcription using data from 420 donors. We find: (a) 7741 cis-eQTLs in islets with a replication rate across 44 GTEx tissues between 40% and 73%; (b) marked overlap between islet cis-eQTL signals and active regulatory sequences in islets, with reduced eQTL effect size observed in the stretch enhancers most strongly implicated in GWAS signal location; (c) enrichment of islet cis-eQTL signals with T2D risk variants identified in genome-wide association studies; and (d) colocalization between 47 islet cis-eQTLs and variants influencing T2D or glycemic traits, including DGKB and TCF7L2. Our findings illustrate the advantages of performing functional and regulatory studies in disease relevant tissues.

The impact of sex on gene expression across human tissues.

Many complex human phenotypes exhibit sex-differentiated characteristics. However, the molecular mechanisms underlying these differences remain largely unknown. We generated a catalog of sex differences in gene expression and in the genetic regulation of gene expression across 44 human tissue sources surveyed by the Genotype-Tissue Expression project (GTEx, v8 release). We demonstrate that sex influences gene expression levels and cellular composition of tissue samples across the human body. A total of 37% of all genes exhibit sex-biased expression in at least one tissue. We identify cis expression quantitative trait loci (eQTLs) with sex-differentiated effects and characterize their cellular origin. By integrating sex-biased eQTLs with genome-wide association study data, we identify 58 gene-trait associations that are driven by genetic regulation of gene expression in a single sex. These findings provide an extensive characterization of sex differences in the human transcriptome and its genetic regulation.

Cell type-specific genetic regulation of gene expression across human tissues.

The Genotype-Tissue Expression (GTEx) project has identified expression and splicing quantitative trait loci in cis (QTLs) for the majority of genes across a wide range of human tissues. However, the functional characterization of these QTLs has been limited by the heterogeneous cellular composition of GTEx tissue samples. We mapped interactions between computational estimates of cell type abundance and genotype to identify cell type-interaction QTLs for seven cell types and show that cell type-interaction expression QTLs (eQTLs) provide finer resolution to tissue specificity than bulk tissue cis-eQTLs. Analyses of genetic associations with 87 complex traits show a contribution from cell type-interaction QTLs and enables the discovery of hundreds of previously unidentified colocalized loci that are masked in bulk tissue.

Dietary metabolite profiling brings new insight into the relationship between nutrition and metabolic risk: An IMI DIRECT study.

BACKGROUND: Dietary advice remains the cornerstone of prevention and management of type 2 diabetes (T2D). However, understanding the efficacy of dietary interventions is confounded by the challenges inherent in assessing free living diet. Here we profiled dietary metabolites to investigate glycaemic deterioration and cardiometabolic risk in people at risk of or living with T2D. METHODS: We analysed data from plasma collected at baseline and 18-month follow-up in individuals from the Innovative Medicines Initiative (IMI) Diabetes Research on Patient Stratification (DIRECT) cohort 1 n = 403 individuals with normal or impaired glucose regulation (prediabetic) and cohort 2 n = 458 individuals with new onset of T2D. A dietary metabolite profile model (Tpred) was constructed using multivariable regression of 113 plasma metabolites obtained from targeted metabolomics assays. The continuous Tpred score was used to explore the relationships between diet, glycaemic deterioration and cardio-metabolic risk via multiple linear regression models. FINDINGS: A higher Tpred score was associated with healthier diets high in wholegrain (ß=3.36 g, 95% CI 0.31, 6.40 and ß=2.82 g, 95% CI 0.06, 5.57) and lower energy intake (ß=-75.53 kcal, 95% CI -144.71, -2.35 and ß=-122.51 kcal, 95% CI -186.56, -38.46), and saturated fat (ß=-0.92 g, 95% CI -1.56, -0.28 and ß=-0.98 g, 95% CI -1.53, -0.42 g), respectively for cohort 1 and 2. In both cohorts a higher Tpred score was also associated with lower total body adiposity and favourable lipid profiles HDL-cholesterol (ß=0.07 mmol/L, 95% CI 0.03, 0.1), (ß=0.08 mmol/L, 95% CI 0.04, 0.1), and triglycerides (ß=-0.1 mmol/L, 95% CI -0.2, -0.03), (ß=-0.2 mmol/L, 95% CI -0.3, -0.09), respectively for cohort 1 and 2. In cohort 2, the Tpred score was negatively associated with liver fat (ß=-0.74%, 95% CI -0.67, -0.81), and lower fasting concentrations of HbA1c (ß=-0.9 mmol/mol, 95% CI -1.5, -0.1), glucose (ß=-0.2 mmol/L, 95% CI -0.4, -0.05) and insulin (ß=-11.0 pmol/mol, 95% CI -19.5, -2.6). Longitudinal analysis showed at 18-month follow up a higher Tpred score was also associated lower total body adiposity in both cohorts and lower fasting glucose (ß=-0.2 mmol/L, 95% CI -0.3, -0.01) and insulin (ß=-9.2 pmol/mol, 95% CI -17.9, -0.4) concentrations in cohort 2. INTERPRETATION: Plasma dietary metabolite profiling provides objective measures of diet intake, showing a relationship to glycaemic deterioration and cardiometabolic health. FUNDING: This work was supported by the Innovative Medicines Initiative Joint Undertaking under grant agreement no. 115,317 (DIRECT), resources of which are composed of financial contribution from the European Union's Seventh Framework Programme (FP7/2007-2013) and EFPIA companies.

Predicting and elucidating the etiology of fatty liver disease: A machine learning modeling and validation study in the IMI DIRECT cohorts.

BACKGROUND: Non-alcoholic fatty liver disease (NAFLD) is highly prevalent and causes serious health complications in individuals with and without type 2 diabetes (T2D). Early diagnosis of NAFLD is important, as this can help prevent irreversible damage to the liver and, ultimately, hepatocellular carcinomas. We sought to expand etiological understanding and develop a diagnostic tool for NAFLD using machine learning. METHODS AND FINDINGS: We utilized the baseline data from IMI DIRECT, a multicenter prospective cohort study of 3,029 European-ancestry adults recently diagnosed with T2D (n = 795) or at high risk of developing the disease (n = 2,234). Multi-omics (genetic, transcriptomic, proteomic, and metabolomic) and clinical (liver enzymes and other serological biomarkers, anthropometry, measures of beta-cell function, insulin sensitivity, and lifestyle) data comprised the key input variables. The models were trained on MRI-image-derived liver fat content (<5% or ≥5%) available for 1,514 participants. We applied LASSO (least absolute shrinkage and selection operator) to select features from the different layers of omics data and random forest analysis to develop the models. The prediction models included clinical and omics variables separately or in combination. A model including all omics and clinical variables yielded a cross-validated receiver operating characteristic area under the curve (ROCAUC) of 0.84 (95% CI 0.82, 0.86; p < 0.001), which compared with a ROCAUC of 0.82 (95% CI 0.81, 0.83; p < 0.001) for a model including 9 clinically accessible variables. The IMI DIRECT prediction models outperformed existing noninvasive NAFLD prediction tools. One limitation is that these analyses were performed in adults of European ancestry residing in northern Europe, and it is unknown how well these findings will translate to people of other ancestries and exposed to environmental risk factors that differ from those of the present cohort. Another key limitation of this study is that the prediction was done on a binary outcome of liver fat quantity (<5% or ≥5%) rather than a continuous one. CONCLUSIONS: In this study, we developed several models with different combinations of clinical and omics data and identified biological features that appear to be associated with liver fat accumulation. In general, the clinical variables showed better prediction ability than the complex omics variables. However, the combination of omics and clinical variables yielded the highest accuracy. We have incorporated the developed clinical models into a web interface (see: https://www.predictliverfat.org/) and made it available to the community. TRIAL REGISTRATION: ClinicalTrials.gov NCT03814915.