Unsupervised Approaches to Placental Protein Clustering: Which Best Captures Signals Linked to Childhood Metabolic Health?

Unsupervised Approaches to Placental Protein Clustering: Which Best Captures Signals Linked to Childhood Metabolic Health? | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Unsupervised Approaches to Placental Protein Clustering: Which Best Captures Signals Linked to Childhood Metabolic Health? Ellen C. Francis, Kristen E. Boyle, Lauren E. Gyllenhammer, Dana Dabelea, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7926327/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 11 You are reading this latest preprint version Abstract Background Placental signaling pathways regulate nutrient transport and fetal growth, with potential long-term consequences for offspring metabolic health. Most prior human studies have focused on individual placental markers, limiting insight into the role of coordinated activity across multiple pathways in relation to offspring outcomes. Objective To compare three unsupervised data reduction techniques for characterizing placental signaling patterns across multiple pathways and assess their associations with neonatal and early childhood adiposity and metabolic biomarkers. Design: Among 108 mother-child pairs from the Healthy Start cohort, we quantified 33 placental signaling proteins and their phosphorylated-to-total protein ratios involved in nutrient sensing, insulin/growth factor signaling, stress/inflammation, and mitochondrial biogenesis using Simple Western assays of term placental villus tissue. We applied consensus clustering, weighted gene correlation network analysis (WGCNA), and principal component analysis (PCA) to derive signaling scores. Model performance (AIC, R², and RMSE) was compared, and associations with offspring outcomes at age 4 years (%fat mass; fasting adiponectin, leptin, insulin, glucose, and lipids) were estimated using multivariable linear regression adjusted for offspring age, race and ethnicity, and maternal pre-pregnancy BMI. Results Consensus clustering outperformed PCA and WGCNA based on model fit statistics. The mTOR/AMPK cluster, characterized by activation of mTOR complex 1 and energy sensing (e.g., phosphorylated 4E-BP1, RPS6, AMPK), was inversely associated with childhood %fat mass (β: − 2.51%, 95% CI: − 4.44, − 0.58). The IGF/Mitochondrial Biogenesis cluster was positively associated with childhood triglyceride levels (17.90 [6.14, 29.60] mg/dL). Conclusion Consensus clustering provided superior model fit compared to WGCNA and PCA. Placental signaling clusters were associated with childhood adiposity and metabolic markers, supporting the relevance of coordinated placental activity to early metabolic programming in a healthy pregnancy cohort. These findings highlight the utility of unsupervised analytic approaches in placental biology and the potential of early-life placental markers to inform pediatric metabolic disease risk. However, which approach is best for summarizing complex protein data is likely dependent on the data structure, dimensionality, and covariance. placenta maternal-fetal exchange insulin signaling nutrient sensing fetal programming cardiometabolic health data reduction unsupervised analysis pregnancy Figures Figure 1 Figure 2 Background The activation of placental signaling pathways that regulate nutrient transport is central to fetal development and may have lasting implications for metabolic health across the life course.( 1 ) Animal studies have demonstrated that disruptions in placental signaling pathways can impair fetal growth. For example, increased nutrient transport via mTOR activation has been associated with fetal overgrowth, while greater placental stress activation (e.g., JNK) has been observed in intrauterine growth restriction—a condition linked to increased fat mass and a less favorable cardiometabolic profile later in life.( 2 – 5 ) Importantly, poor cardiometabolic traits in childhood track across the life course and are predictors of future obesity,( 6 – 10 ) type 2 diabetes, and metabolic syndrome,( 11 – 15 ) highlighting the importance of early identification of biological risk before overt complications arise.( 16 ) In human pregnancies, variation in the above placental signaling pathways have also been associated with neonatal and childhood metabolic outcomes, including adiposity, lipid profiles, and insulin sensitivity.( 17 – 24 ) However, much of this work has focused on discrete proteins or single pathways, often in the context of maternal complications such as obesity or gestational diabetes (GDM).( 25 ) As a result, there is limited understanding of how the interrelationship among these pathways in healthy, low-risk pregnancies—where subtle signaling variation may still be biologically meaningful. As biological pathways operate in concert, understanding their coordination may be particularly important in normal low-risk pregnancies where severe dysregulation of a single pathway is likely to be absent. These subtle shifts may nonetheless influence long-term health trajectories, given that gestation is a sensitive period for developmental programming. Advances in omics science have enabled the systematic measurement of multiple biological pathways concurrently. In parallel, analytic frameworks such as clustering, network analyses, and data reduction have been developed to identify patterns of co-expression or co-regulation within high-dimensional datasets. These unsupervised approaches can reveal latent structure across interacting pathways, offering a systems-level view of molecular signaling that complements traditional single-protein models.( 26 , 27 ) Despite their utility for summarizing large amounts of data, we are not aware of studies that have applied or compared such methods to classically measured protein data, such as those obtained via Western assays. Our primary objective was to compare three commonly used unsupervised analytic methods— consensus clustering( 28 ), weighted gene correlation network analysis (WGCNA)( 29 ), and principal component analysis (PCA)—for identifying placental protein signaling patterns potentially relevant to fetal and childhood metabolic outcomes. We applied these methods to a panel of placental proteins quantified using Simple Western Assays (WES) and evaluated their performance in models predicting neonatal and early childhood adiposity. Secondary analyses examined associations with broader childhood cardiometabolic health outcomes. These three methods were compared because PCA one of the most used approaches for data reduction, WGCNA is a commonly used network-based technique, and consensus clustering includes internal resampling to identify cluster stability, which is not a standard step in PCA or WGCNA. These methods summarize patterns of co-expression or shared variance across multiple proteins, capturing broader signaling relationships across multiple pathways. Methods Participants and sample collection We used data from the Healthy Start Study (ClinicalTrials.gov; NCT02273297). The Healthy Start pre-birth cohort study enrolled 1,410 women aged ≥ 16 years of age at < 24 weeks of gestation from prenatal clinics at the University of Colorado Hospital between 2009–2014. We excluded women with prior diabetes, a history of prior preterm birth < 25 gestational weeks or fetal death, asthma with active steroid management, serious psychiatric illness or multiple gestation.( 30 ) As part of the ancillary Healthy Start BabyBUMP Project we collected and snap froze trophoblast villi samples from placentas after delivery in a convenience subsample (N = 111) of the Healthy Start cohort. Of these women, we excluded those women with a pre-term delivery (N = 2) and who were completely missing offspring outcomes (N = 1). The final sample included 108 placentas. Of the 108 mother-offspring pairs, all were invited to complete a follow-up research visit in early childhood (mean offspring age 4.8 ± 0.6), and 67% (n = 72) of those children returned for an in-person visit during which weight and height were measured (details in proceeding sections). Among the children who returned for the early childhood visit, 72% (n = 52) also completed a blood collection. Placental samples and signaling The placental proteins were selected a priori based on their roles in insulin/growth factor signaling, inflammation, nutrient transport, and energy sensing.( 19 , 31 ) We homogenized ~ 20mg of frozen trophoblast placental villus tissue in 75µL ice-cold buffer D (250mM sucrose, 10mM HEPES, pH 7.4) that had a 1:100 dilution of protease and phosphatase inhibitors. Next, we used Simple Western Assays with WES (ProteinSimple, Santa Clara, CA) to measure the phosphorylation and total abundance of the following proteins: ribosomal protein S6 (RPS6), p70 ribosomal protein S6 kinase 1 (S6K1), eukaryotic translation initiation factor 4E-binding protein 1 (4E-BP1), protein kinase B (Akt), protein kinase C-α (PKCα), insulin receptor β (IRβ), insulin-like growth factor 1 receptor (IGF-1r), glycogen synthase kinase-3 beta (GSK3β), extracellular signal-regulated kinase (ERK), 1/2 pro-caspase 1, signal transducer and activator of transcription 3 (STAT3), c-Jun N-terminal kinase 1 and 2 (JNK1 and JNK2), interleukin-1 beta (IL-1β), p38 mitogen-activated protein kinase (p38 MAPK), AMP-activated protein kinase (AMPK), eukaryotic initiation factor 2 alpha (eIF2α), O-linked N-acetylglucosamine transferase (OGT), phosphoinositide 3-kinase (PI3K), 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1), and peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α). We used a 0.1 mg/mL total protein concentration and ran the WES plates according to the manufacturer’s instructions, with slight modification (200V, 55m separation time). We included an equalizer sample on each plate with clean median values for each protein to control for batch variation. We also multiplexed a loading control (vinculin or β-actin) in each capillary and normalized the protein levels to it. Maternal and child data Maternal lifestyle and clinical data Maternal race and ethnicity, educational attainment, parity, and smoking status during pregnancy were self-reported via questionnaire. GDM status was abstracted from medical records and all data collection are reported in detail elsewhere.( 32 , 33 ) We calculated maternal pre-pregnancy BMI based on pre-pregnancy weight obtained from medical records (89%) or self-report (11%), and measured height at the first research visit. Gestational weight gain (GWG) was estimated by subtracting pre-pregnancy weight from the last clinically measured weight during pregnancy and categorized based on the Institute of Medicine 2009 Guidelines.( 34 ) Beginning in the 1st trimester, maternal diet was assessed via the Automated Self-Administered 24-h Dietary Recall (ASA24).( 35 ) We calculated the Healthy Eating Index-2010 (HEI)( 33 , 36 – 38 ) to capture diet quality in accordance with the 2010 Dietary Guidelines for Americans. ( 39 , 40 ) Child metabolic outcomes Offspring fat mass and fat free mass at birth and in early childhood were measured using whole body air displacement plethysmography (ADP; PeaPod and BodPod, Life Measurement, Inc.) with the Pediatric Option. ( 41 ) Measurements for each participant were taken in triplicate and the average of the two closest measures was used for analyses. The children’s height and weight were measured by trained nurses. Age-specific body mass index (BMI) percentile were calculated according to the World Health Organization (WHO) growth reference ( 42 , 43 ). Fasting serum was used to measure triglycerides (TGs), total cholesterol, high-density lipoprotein (HDL), low-density lipoprotein (LDL), and glucose using manufacturer prepackaged enzymatic kits and the AU400e Chemistry Analyzer (Olympus America). Insulin was measured using a radioimmune assay, and leptin and adiponectin were measured using a Multiplex assay kit, all by Millipore Corporation. Child lifestyle data Early childhood dietary data was collected via two Automated Self-Administered 24-hour Dietary Assessment Tool recalls (1 weekend and 1 weekday, with caretaker proxy) ( 44 ). Nutrient and caloric intake were derived using Nutrition Data System for Research software package. Offspring physical activity was measured using wGT3X-BT ActiGraph accelerometers (Pensacola, FL) worn for 7 days during waking hours on the waist.( 45 ) Average weekly moderate to vigorous physical activity was categorized as adequate (≥ 1 hour/day) using intensity cut points established in youth.( 46 , 47 ) We derived the child’s HEI scores using the same procedure as for the mothers, except the 2015 U.S. Dietary Guidelines were used to align with timing of data collection. Data and Statistical Analysis The analytical workflow is shown in Fig. 1 . All data and statistical analysis were performed using R Statistical Software (v. 4.0.0 R Core Team 2024). Please note that full details on data and statistical analyses are in Supplemental Methods . A brief overview is provided below. The analysis proceeded in five sequential steps. ( 1 ) Placental protein data were preprocessed by normalizing expression levels to total placental protein and calculating phospho-to-total protein ratios to capture pathway activation. Missing values were imputed using multiple imputation with chained equations (MICE) based on classification and regression trees. ( 2 ) To derive latent variables summarizing placental signaling, we applied three unsupervised data reduction methods—consensus clustering, weighted gene co-expression network analysis (WGCNA), and principal component analysis (PCA)—each generating summarized scores from protein expression profiles. ( 3 ) To evaluate model performance, we tested associations between the reduced dimensions and offspring fat mass percentage (neonatal and early childhood) using multivariable linear regression adjusted for child sex and age. Model fit was assessed using Akaike information criterion (AIC), adjusted R², root mean square error (RMSE), paired t-tests with bootstrap resampling, and Akaike weights. ( 4 ) The best-performing method was selected based on its ability to explain variation in offspring fat mass while balancing fit and parsimony. ( 5 ) Summary scores from the selected data reduction method were used in adjusted models to examine associations with early childhood fat mass and cardiometabolic outcomes. All models were evaluated using both complete-case and multiply imputed data Placenta Data Preprocessing Placental protein expression was normalized to total protein levels, and phosphorylated-to-total ratios were computed to capture signaling activation. Total protein expression values were also retained to assess co-expression patterns. A small number of samples with partial missingness were imputed using multiple imputation. Full preprocessing and imputation details are provided in Supplemental Methods Section 1. Data Reduction Methods Three unsupervised approaches were applied to reduce dimensionality of the placental signaling data: consensus clustering, WGCNA, and PCA. Each method generated latent variables summarizing placental protein patterns for downstream modeling. We retained dimensions based on established criteria for each method (e.g., cluster stability, module size, eigenvalue thresholds). See Supplemental Methods Section 2 for technical implementation details. Model Comparison and Main Analyses To compare the utility of each data reduction approach and to select the optimal method for downstream analysis, we examined associations of the latent structured generated by each with neonatal and early childhood fat mass percentage using multivariable linear regression adjusted for child age and sex. Model performance was evaluated using Akaike Information Criterion (AIC), adjusted R², and root mean squared error (RMSE). The best-performing method was selected based on consistent performance across these metrics (see Supplemental Methods Section 3 ). Association of Placental Clusters and Child Cardiometabolic Outcomes Following identification of the optimal dimension reduction approach, we investigated associations of the optimal method corresponding latent structure scores with offspring cardiometabolic outcomes. Here, we specified three sequentially adjusted multivariable models using complete case (CC) and Multivariate Imputation by Chained Equations (MICE) ( Supplemental Methods Section 4) . Model 1 adjusted for offspring sex and age at follow-up. Model 2 adjusted for Model 1 and observed gestational weight gain and pre-pregnancy BMI as confounders of placenta-child outcome association that is not explained by the placental signaling pathways measured in our study (e.g., uteroplacental blood flow). Model 3 adjusted for Model 1 and child diet quality (total HEI score) and physical activity (minutes of moderate-to-vigorous activity) measured at follow-up to account for differences in children’s lifestyle (e.g., precision covariates). We used several criteria to evaluate the robustness of associations between placental signaling clusters and child metabolic outcomes. First, estimates had to demonstrate consistency across CC and MICE, as well as across the different covariate adjustment models, in terms of both magnitude and direction. Second, we considered associations statistically significant at a nominal alpha of 0.05, corresponding to a 95% confidence interval that excluded the null. Additionally, given the relatively small sample size and number of comparisons made, we supplemented interpretation of results with estimates that met statistical significance at alpha of 0.15 Sensitivity analysis We performed several sensitivity analyses in the data reduction and association testing phases. First, although we include R 2 in our assessment of the different data reduction techniques, we further assessed if variance explained in the outcome was due to differences in the number of dimensions for each reduction technique (consensus clustering = 4, WGCNA = 5, and PCA = 9). The top three dimensions were chosen based on the 1st three PCs, the three WGCNA modules with the highest mean connectivity, and the three largest consensus clusters. Second, we specified joint models that included the linear combination of all placental consensus clustering scores with neonatal and childhood fat mass percentage as separate outcomes and evaluated the estimates for consistency with the series of individual models where each cluster was tested separately. Lastly, we compared differences in mean cluster scores by maternal characteristics to identify upstream factors associated with placental signaling clusters. Results Characteristics Maternal and child characteristics are presented in Table 1 . Most of the pregnancies were from non-Hispanic white females who had completed at least some college. The mean pre-pregnancy BMI was 24.6 ± 4.2 kg/m 2 and 10% of women developed GDM. The mean gestational age of offspring was at term (277 ± 8.5 days), birth weight-for-gestational age percentile of 29.9 (24.5), and 51.9% were males. At the follow-up research visit, the mean age of children was 4.79 ± 0.6 years, BMI percentile was 46.2 ± 22.4, HEI score was 60.8 ± 10.1, and minutes of vigorous/moderate activity per week was 48.5 ± 25.3. Table 1 Maternal and offspring characteristics Characteristics; mean (SD), n (%) Overall N = 108 Maternal Age at index pregnancy (years) 29.1 (6.1) Race-ethnicity Non-Hispanic White 67 (62.0%) Non-Hispanic Black 12 (11.1%) Hispanic 23 (21.3%) Non-Hispanic Other 6 (5.6%) Education College Degree 26 (24.1%) Grad Degree 29 (26.9%) Some college 25 (23.1%) HS or less 28 (25.9%) Pre-pregnancy BMI (kg/m 2 ) 25.0 (5.6) Gestational diabetes mellitus Yes 5 (4.6%) Total HEI score 56.8 (13.8) Total gestational weight gain (lbs) 13.5 (7.7) Smoked during pregnancy 8 (7.4%) Neonatal Male sex 56 (51.9%) Gestational age at delivery (days) 277 (8.5) Weight-for-gestational age percentile b 29.9 (24.5) Childhood Age (years) 4.79 (0.6) Healthy Eating Index (score) 60.8 (10.1) Vigorous/Moderate activity per week (minutes) 48.5 (25.3) BMI, (kg/m 2 ) c 15.2 (1.23) BMI, (percentile) 46.2 (22.4) a Other includes, Asian, American Indian/Alaska natives, Hawaiian/Pacific Islanders b Birthweight z-score specific to gestational age derived from U.S. natality reference ( 48 ) c BMI specific to age derived from WHO growth standards ( 49 ) HEI; Healthy Eating Index Data reduction and protein clustering The placental protein data showed a modest correlation structure, with a median pairwise Spearman correlation of 0.13 (IQR: 0.06–0.22). The results of the three unspurivsed data reduction methods, consensus clustering, WGCNA, and PCA, are shown in Fig. 2. Across the three methods, several consistent patterns emerged. A subset of proteins involved in growth and metabolic signaling pathways, including IGF1r, AKT, and PGC1α, frequently appeared together. Proteins associated with cellular stress or inflammatory signaling, such as total levels of STAT3, eIF2a, p38MAPK, and JNK also clustered together or loaded onto similar components. The grouping of these proteins were often combined with their phophorylated forms in growth and metabolic signaling such as JNK ratios and ERK1/2 ratios. In addition, activation of proteins involved in mTORC1 regulating translation, including phosphorylation of 4E-BP1 and RPS6, frequently appeared together across methods. Across both child adiposity outcomes, models using all available dimensions showed that consensus clustering consistently provided the best model fit (lowest AIC), the highest explanatory power (adjusted R²), and the strongest model support (Akaike weights). For instance, when evaluating neonatal fat mass percent, consensus cluster metrics (AIC = 589.5; Akaike weight = 0.978) indicated it was more likely to be the best model compared to PCA (AIC = 600.5; weight = 0.004) and WGCNA (AIC = 597.5; weight = 0.018). Further, the RMSE was significantly lower for the consensus clustering model compared to WGCNA (p < 2.2e-16), though not significantly different compared to PCA (p = 0.18) ( Supplemental Table 1 ). Given that consensus clustering demonstrated better model performance more consistently, we used consensus clusters and their summary scores as the main predictor. The four consensus clusters were labeled based on the proteins grouped into each cluster. “ mTOR/AMPK ” (n = 4 proteins) included the phosphorylated proteins from insulin/IGF signaling and energy-sensing pathways AMPK ratio and RPS6 ratio. “ IGF/Mitochondrial Biogensis ” (n = 2 proteins) contained IGF1r and PGC1α which are in insulin signlaing and mitochondrial biogenesis pathways. “ Placental Insulin Coordination ” (n = 15 proteins) which was the largest and comprised primarily of phosphorylated proteins from all pathways except mitochondrial biogenesis and cortisol metabolism. “ Inflammation/Stress ” (n = 12 proteins) was comprised primarily of total protein concentration markers of MAPK signaling, inflammatory signaling, and stress response, such as ERK1/2, p38MAPK, STAT3, JNK1, and IL-1β. Cluster scores represent the mean standardized (z-scored) expression of proteins within each cluster for each participant, such that higher scores indicate relatively higher expression or activation of proteins assigned to that cluster compared to the overall mean in our sample of participants. Associations of placental protein clusters with child cardiometabolic outcomes Several consensus clusters showed consistent associations with childhood cardiometabolic outcomes across adjustments for child age and sex (Model 1), maternal pre-pregnancy BMI and gestational weight gain (Model 2; Table 2 ), and child lifestyle factors (Model 3; Supplemental Table 3 ). Here we highlight patterns of association that showed similar directions of effect across model adjustment, recognizing that in a few instances, estimates varied in magnitude and precision between CC and MICE. In both CC and MICE models, a higher “ mTOR/AMPK ” cluster score was associated with lower childhood fat mass percentage (e.g., β = − 2.51%, 95% CI: − 4.44, − 0.58; p = 0.01 in the MICE Model 2). This association was slightly attenuated after adjusting for concurrent lifestyle. Higher “ IGF/Mitochondrial Biogenesis ” cluster scores were consistently associated with higher triglycerides in the CC models (e.g., β = 17.90 mmol/l, 95% CI: 6.14, 29.60; p < 0.001 in CC Model 2), with reduced precision and attenuation in MICE models (e.g., β = 8.20 mmol/l, 95% CI: − 1.73, 18.13; p = 0.11 in MICE Model 2). “ Placental Insulin Coordination ” scores trended with lower insulin levels, though confidence intervals crossed zero (e.g., β = − 2.63 pmol/l, 95% CI: − 5.63, 0.37; p = 0.09 in MICE Model 2). Higher “ Inflammation/Stress ” scores were associated with lower LDL cholesterol in CC models, with confidence intervals excluding zero after adjusting for concurrent lifestyle (β = − 17.00 mmol/l, 95% CI: − 33.30, − 0.67; p = 0.04 in CC Model 3). This association was attenuated and became imprecise in all MICE models. Table 2 Association of Placental Signaling Clusters with Childhood Metabolic Health Indicators MODEL 1 (age & sex) MODEL 2 (Model 1 + Maternal pre-pregnancy BMI) Cluster, Outcome CC MICE N BETA (85% CI) P N BETA (85% CI) P N BETA (85% CI) P N BETA (85% CI) P mTOR/AMPK Glucose, mmol/l 40 0.54 (-1.35, 2.43) 0.68 72 0.38 (-1.96, 2.72) 0.82 40 0.51 (-1.23, 2.25) 0.67 72 0.45 (-1.86, 2.77) 0.78 Insulin, pmol/l 52 -0.32 (-1.53, 0.89) 0.70 72 -0.24 (-1.30, 0.81) 0.74 52 -0.32 (-1.54, 0.90) 0.74 72 -0.24 (-1.30, 0.82) 0.74 HOMA-IR 36 -0.04 (-0.43, 0.34) 0.87 72 -0.07 (-0.40, 0.26) 0.77 36 -0.05 (-0.44, 0.34) 0.85 72 -0.06 (-0.39, 0.27) 0.79 Adiponectin, ug/ml 51 -2.16 (-4.38, 0.06) 0.16 72 -2.12 (-4.46, 0.22) 0.19 51 -2.11 (-4.34, 0.11) 0.22 72 -2.18 (-4.52, 0.17) 0.18 Leptin, µg/l 50 -0.37 (-1.21, 0.48) 0.53 72 -0.03 (-0.83, 0.77) 0.96 50 -0.38 (-1.23, 0.47) 0.44 72 -0.04 (-0.84, 0.77) 0.94 Cholesterol, mmol/l 48 3.11 (-3.30, 9.51) 0.48 72 1.24 (-5.39, 7.88) 0.79 48 3.19 (-3.22, 9.59) 0.49 72 1.35 (-5.29, 7.98) 0.77 Triglycerides, mmol/l 48 0.58 (-7.46, 8.62) 0.92 72 -2.30 (-10.89, 6.29) 0.70 48 0.48 (-7.57, 8.53) 0.90 72 -2.37 (-10.95, 6.21) 0.69 HDL, mmol/l 48 1.56 (-1.97, 5.09) 0.52 72 0.93 (-3.34, 5.20) 0.75 48 1.63 (-1.85, 5.11) 0.53 72 1.02 (-3.23, 5.27) 0.73 LDL, mmol/l 37 -5.07 (-12.07, 1.92) 0.29 72 -3.59 (-9.38, 2.20) 0.37 37 -5.04 (-12.14, 2.07) 0.31 72 -3.59 (-9.42, 2.23) 0.37 %Fat mass 67 -2.44 (-3.85, -1.02) 0.01* 72 -2.44 (-3.83, -1.04) 0.01* 67 -2.44 (-3.87, -1.01) 0.01* 72 -2.44 (-3.85, -1.03) 0.01* BMI percentile 72 -1.77 (-7.15, 3.61) 0.63 72 -1.77 (-7.08, 3.55) 0.63 72 -1.86 (-7.26, 3.54) 0.53 72 -1.86 (-7.20, 3.47) 0.62 IGF/Mitochondrial Biogenesis Glucose, mmol/l 40 0.81 (-0.75, 2.38) 0.45 72 0.86 (-1.27, 2.99) 0.56 40 0.76 (-0.68, 2.21) 0.49 72 0.75 (-1.37, 2.86) 0.61 Insulin, pmol/l 52 0.22 (-1.04, 1.48) 0.80 72 0.42 (-0.54, 1.37) 0.53 52 0.22 (-1.05, 1.49) 0.77 72 0.42 (-0.54, 1.38) 0.53 HOMA-IR 36 -0.04 (-0.39, 0.32) 0.88 72 -0.07 (-0.36, 0.21) 0.71 36 -0.03 (-0.38, 0.33) 0.92 72 -0.08 (-0.37, 0.21) 0.68 Adiponectin, ug/ml 51 1.64 (-0.51, 3.79) 0.27 72 0.98 (-1.33, 3.29) 0.54 51 1.63 (-0.52, 3.79) 0.23 72 1.07 (-1.24, 3.38) 0.51 Leptin, µg/l 50 0.68 (-0.36, 1.72) 0.34 72 0.36 (-0.74, 1.45) 0.64 50 0.68 (-0.37, 1.73) 0.36 72 0.37 (-0.71, 1.45) 0.62 Cholesterol, mmol/l 48 1.70 (-3.35, 6.74) 0.63 72 1.07 (-3.67, 5.82) 0.74 48 1.33 (-3.76, 6.41) 0.55 72 0.92 (-3.82, 5.66) 0.78 Triglycerides, mmol/l 48 11.62 (5.85, 17.39) 0.01* 72 8.34 (1.83, 14.84) 0.07~ 48 12.25 (6.51, 18.00) 0.01* 72 8.48 (1.88, 15.08) 0.06~ HDL, mmol/l 48 0.72 (-2.06, 3.50) 0.71 72 0.78 (-2.74, 4.29) 0.75 48 0.40 (-2.36, 3.17) 0.48 72 0.64 (-2.86, 4.14) 0.79 LDL, mmol/l 37 2.20 (-3.71, 8.12) 0.59 72 -0.77 (-6.45, 4.92) 0.85 37 2.26 (-3.75, 8.27) 0.61 72 -0.77 (-6.49, 4.95) 0.85 %Fat mass 67 0.53 (-0.76, 1.81) 0.55 72 0.53 (-0.74, 1.80) 0.55 67 0.53 (-0.77, 1.83) 0.65 72 0.53 (-0.75, 1.81) 0.55 BMI percentile 72 3.00 (-1.62, 7.63) 0.35 72 3.00 (-1.57, 7.57) 0.34 72 3.17 (-1.48, 7.81) 0.37 72 3.17 (-1.42, 7.76) 0.32 Placental Insulin Coordination Glucose, mmol/l 40 -0.15 (-4.07, 3.76) 0.95 72 -2.80 (-6.68, 1.07) 0.30 40 -1.56 (-5.22, 2.11) 0.58 72 -3.79 (-7.58, -0.01) 0.15 Insulin, pmol/l 52 -3.01 (-5.71, -0.30) 0.11~ 72 -2.31 (-4.48, -0.14) 0.12~ 52 -3.02 (-5.78, -0.27) 0.10~ 72 -2.40 (-4.62, -0.18) 0.12~ HOMA-IR 36 -0.97 (-1.81, -0.13) 0.10~ 72 -0.22 (-1.47, 1.04) 0.80 36 -1.10 (-1.95, -0.26) 0.06~ 72 -0.29 (-1.53, 0.96) 0.74 Adiponectin, ug/ml 51 1.59 (-3.82, 7.00) 0.67 72 2.47 (-2.84, 7.79) 0.50 51 2.13 (-3.34, 7.59) 0.70 72 3.19 (-2.37, 8.75) 0.41 Leptin, µg/l 50 -0.89 (-2.78, 1.01) 0.50 72 -0.65 (-2.52, 1.23) 0.62 50 -0.75 (-2.70, 1.20) 0.65 72 -0.57 (-2.56, 1.42) 0.68 Cholesterol, mmol/l 48 10.25 (-2.24, 22.74) 0.24 72 8.07 (-3.34, 19.48) 0.31 48 9.43 (-3.17, 22.03) 0.38 72 7.23 (-4.25, 18.70) 0.36 Triglycerides, mmol/l 48 2.79 (-13.05, 18.63) 0.80 72 9.26 (-14.06, 32.58) 0.57 48 3.97 (-11.97, 19.91) 0.51 72 10.45 (-12.81, 33.72) 0.52 HDL, mmol/l 48 2.67 (-4.29, 9.64) 0.58 72 -0.15 (-8.89, 8.60) 0.98 48 1.93 (-4.99, 8.85) 0.93 72 -1.18 (-9.95, 7.59) 0.85 LDL, mmol/l 37 1.36 (-15.97, 18.68) 0.91 72 3.65 (-12.35, 19.65) 0.74 37 1.89 (-15.88, 19.67) 0.85 72 3.80 (-12.39, 20.00) 0.74 %Fat mass 67 2.19 (-0.69, 5.07) 0.27 72 2.19 (-0.65, 5.04) 0.27 67 2.27 (-0.68, 5.23) 0.19 72 2.27 (-0.65, 5.19) 0.26 BMI percentile 72 9.85 (-0.66, 20.36) 0.18 72 9.85 (-0.54, 20.24) 0.17 72 11.35 (0.64, 22.06) 0.08~ 72 11.35 (0.77, 21.94) 0.12~ Inflammation/Stress Glucose, mmol/l 40 0.17 (-2.60, 2.93) 0.93 72 -0.24 (-4.45, 3.98) 0.94 40 -0.51 (-3.08, 2.06) 0.71 72 -0.59 (-4.84, 3.66) 0.84 Insulin, pmol/l 52 1.50 (-0.63, 3.63) 0.31 72 1.22 (-0.56, 3.01) 0.33 52 1.56 (-0.62, 3.73) 0.27 72 1.23 (-0.58, 3.04) 0.33 HOMA-IR 36 0.44 (-0.11, 0.99) 0.25 72 0.30 (-0.32, 0.92) 0.49 36 0.40 (-0.16, 0.96) 0.30 72 0.28 (-0.35, 0.91) 0.52 Adiponectin, ug/ml 51 0.21 (-3.86, 4.28) 0.94 72 0.76 (-4.40, 5.92) 0.83 51 0.51 (-3.59, 4.61) 0.74 72 1.02 (-4.20, 6.23) 0.78 Leptin, µg/l 50 -0.90 (-2.38, 0.57) 0.37 72 -0.63 (-2.23, 0.97) 0.57 50 -0.82 (-2.33, 0.68) 0.31 72 -0.59 (-2.24, 1.05) 0.61 Cholesterol, mmol/l 48 -7.34 (-16.51, 1.84) 0.25 72 -7.34 (-15.90, 1.22) 0.22 48 -8.90 (-18.19, 0.39) 0.22 72 -7.87 (-16.54, 0.80) 0.19 Triglycerides, mmol/l 48 -7.45 (-18.97, 4.07) 0.35 72 -4.55 (-17.12, 8.03) 0.60 48 -6.18 (-17.98, 5.61) 0.32 72 -4.26 (-17.02, 8.51) 0.63 HDL, mmol/l 48 0.40 (-4.73, 5.53) 0.91 72 -0.10 (-5.07, 4.87) 0.98 48 -0.63 (-5.79, 4.52) 0.83 72 -0.51 (-5.52, 4.50) 0.88 LDL, mmol/l 37 -12.73 (-22.46, -3.00) 0.06~ 72 -11.24 (-23.55, 1.07) 0.19 37 -12.71 (-22.70, -2.72) 0.06~ 72 -11.33 (-23.86, 1.19) 0.19 %Fat mass 67 -0.05 (-2.63, 2.53) 0.98 72 -0.05 (-2.60, 2.50) 0.98 67 -0.05 (-2.67, 2.57) 0.90 72 -0.05 (-2.63, 2.54) 0.98 BMI percentile 72 -0.33 (-9.46, 8.80) 0.96 72 -0.33 (-9.36, 8.69) 0.96 72 0.11 (-9.09, 9.31) 0.87 72 0.11 (-8.99, 9.20) 0.99 Notes: Missing data were imputed using multiple imputation by chained equations (MICE), generating five imputed datasets. The imputation model included maternal and child predictors, with predictive mean matching for continuous variables and logistic regression for categorical variables. Estimates represent pooled regression coefficients across imputations, with standard errors calculated using Rubin’s rules. Confidence intervals and p-values were derived using a normal approximation (z-test), rather than the default t-distribution approach. Abbreviations: AMPK, AMP-activated protein kinase; mTOR, Mechanistic Target of Rapamycin, IGF, Insulin Like Growth Factor; HOMA-IR, Homeostatic Model Assessment of Insulin Resistance; HDL, high-density lipoprotein; LDL, low-density lipoprotein; TGs, Triglycerides; BMI, body mass index; *P-value <0.05 (95% confidence); ~ P-value<0.15 (85% confidence) Model 1: Age and sex Model 2 (Maternal): pre-pregnancy body mass index (kg/m2) Models of neonatal fat mass % are reported in Supplemental Table 2. Sensitivity analysis In sensitivity analyses the dimensions retained were restricted to the top three consensus clusters based on size (“ Placental Insulin Coordination ”, “ mTOR/AMPK ”, “ Inflammation/Stress ”), the top three PCs based on variance explained (1 through 3), and the top three WGCNA modules based on mean module connectivity (“brown”, “turquoise”, “blue”). In these analyses, PCs performed slightly better for neonatal fat mass percentage; however, for childhood fat mass percentage, consensus clustering remained the best model ( Supplemental Table 1 ). In general, models that mutually adjusted for all four consensus clusters had similar results in association with neonatal and childhood fat mass to those that assessed each cluster individually ( Supplemental Table 2 ), suggesting independence of the pathways captured by each cluster. We utilized separate models for each cluster because our interest was to identify associations with clusters of proteins, without holding constant the activity of the other proteins measured and assigned to different clusters. Discussion In this pregnancy cohort with longitudinal follow-up from birth through early childhood (median offspring age 4.79 ± 0.6 years), we compared three data reduction methods—consensus clustering, WGCNA, and PCA—for capturing placental signaling across multiple pathways potentially involved in in utero programming of offspring cardiometabolic health. Based on model fit metrics, consensus clustering emerged as the optimal for predicting neonatal and early childhood fat mass percentage. The clusters, which captured signaling in pathways of mTOR, AMPK, IGF/insulin signaling, mitochondrial biogenesis, and inflammation and stress responses were also associated with offspring cardiometabolic outcomes at four years of age. In our comparison of the data reduction approaches, the general modest correlation structure observed in the placental protein data (median pairwise Spearman correlation 0.13; IQR: 0.06–0.22) may have influenced the relative performance of the data reduction methods in our analysis. Consensus clustering outperformed PCA and WGCNA in the models using all reduced dimensions, likely due to its ability to detect localized co-variation (e.g., smaller sets of variation) without relying on strong global correlation. Each approach has distinct assumptions about the underlying structure of [biological] data, particularly with respect to correlation. PCA reduces dimensionality by identifying orthogonal axes of shared variance and retains the global correlation structure, allowing each protein to contribute to multiple components.(50) WGCNA also relies on global correlation, using pairwise relationships to identify modules of highly correlated proteins.(29) In contrast, consensus clustering does not explicitly rely on correlation; instead, it uses repeated resampling to identify stable groupings of proteins that co-vary across subsets of samples,(28) which may capture more nuanced groupings of proteins that are co-regulated. While consensus clustering of placental protein data resulted in the best model fit and the highest explanatory power, the best model fit will vary depending on the number and types of proteins measured, as well as the specific outcome. We present these three complementary approaches and their comparisons as a guide for others seeking an alternative to single protein models. Whether consensus clustering is an optimal approach that could be adopted more generally can only be determined with replication in other study populations. We found that the “ mTOR/AMPK ” cluster, characterized by increased activation of RPS6, 4E-BP, and AMPK, was inversely associated with neonatal and early childhood fat mass percentage. This cluster also showed a general inverse pattern with the other metabolic markers during early childhood, although there was imprecision in these models. These overall findings suggest a potential link between this placental signaling pattern and lower offspring adiposity, with possible implications for longer-term metabolic health. These findings align with prior research reporting inverse associations of AMPK activation and birthweight,(51, 52) as well as broader evidence linking AMPK activity with improved metabolic outcomes.(53) Importantly, most prior studies on placental mTOR and AMPK signaling were conducted in high-risk pregnancies (e.g., maternal obesity, hyperglycemia, or hypoxia), where placental function is often pathologically disrupted.(25, 52, 54-56) In contrast, our cohort primarily comprised of low-risk pregnancies, allowed us to observe more physiological—and possibly adaptive—signaling patterns, rather than pathological ones. A particularly notable aspect of the “ mTOR/AMPK ” cluster was the co-activation of AMPK with mTORC1 downstream targets RPS6 and 4E-BP1. This may appear counterintuitive, as AMPK is classically thought to inhibit mTOR signaling, particularly in energy-stressed states.(57) However, placental signaling does not always follow traditional nutrient-sensing paradigms observed in liver, adipose, or skeletal muscle. Indeed, prior animal studies have shown that placental mTOR can be inhibited even in the absence of AMPK activation, suggesting distinct regulatory mechanisms.(22, 58) Co-activation of AMPK and mTOR effectors may reflect a placenta-specific strategy to balance energy sensing with protein synthesis, in the context of supporting both placental and fetal growth.(59) For example, we have previously speculated that 4E-BP1 activation may preferentially support placental growth and biosynthesis, whereas RPS6 signaling may be more directly involved in nutrient transport, oxidative metabolism, and fetal growth.(31) In the context of our study, the clustering of AMPK, 4E-BP1, and RPS6—along with their inverse association with neonatal fat mass— may indicate a coordinated signaling state to balance fetal growth priorities and placental function to prevent excess nutrient transfer to the fetus. However, this remains a hypothesis that requires further mechanistic investigation. Given the complex regulation of placental signaling by several inputs (e.g., glucose, amino acids, blood flow, oxygenation), these data underscore the need for future studies to disentangle how integrated placental signaling networks shape fetal nutrient exposure and influence early metabolic trajectories.(60) We found that “ IGF/Mitochondria Biogenesis ” cluster—marked by higher IGF-1r and PGC1α—was positively associated with childhood triglyceride levels. While adjustment for child’s lifestyle modestly attenuated this association and the confidence intervals included the null, the pattern remained directionally consistent. The clustering of IGF-1r and PGC1α is biologically plausible, as IGF-1r is a known upstream regulator of PGC1α expression and mitochondrial biogenesis.(61) Beyond the role of PGC1α in mitochondrial remodeling, it also influences lipid metabolism including long-chain fatty acid oxidation and hepatic triglyceride storage and secretion.(62, 63) Taken together, these findings suggest that enhanced placental IGF-1 signaling and mitochondrial activity may prime lipid metabolic pathways in utero, potentially increasing susceptibility to elevated triglycerides in childhood. The attenuation of associations after accounting for child lifestyle exposures may reflect an interaction between early programing and postnatal obesogenic environments, potentially amplifying the effects of early programming through a secondary “second-hit” mechanism.(64) We observed that the “ Placental Insulin Coordination” cluster—which included phosphorylated proteins across insulin signaling (IRβ, ERK1/2), mTOR nutrient transport (total RPS6), and stress-response pathways (JNK)—was modestly associated with lower HOMA-IR in early childhood. These associations were not attenuated by adjustment for maternal BMI or child lifestyle. While the clustering of activated proteins from these different pathways may seem inconsistent with canonical nutrient-sensing models, the clustering pattern may represent coordinated crosstalk among insulin, nutrient-sensing, and stress-response pathways that maintain placental homeostasis in low-risk pregnancies. Mechanistically, insulin receptor signaling activates both mitogenic (ERFK1/2) and metabolic (AKT) arms, with feedback via mTOR and JNK that modulate the intensity of downstream signals.(65, 66) As such, the clustering of activated ERK1/2 and JNK with total RPS6 may indicate a more responsive and balanced placental signaling state—potentially enhancing fetal metabolic programing and contributing to improved insulin sensitivity in offspring. The “ Inflammation/Stress ” cluster, comprised of proteins from MAPK, inflammatory signaling, and stress response signaling (ERK1/2, p38MAPK, STAT3, JNK1, and IL-1β), was inversely associated with childhood LDL cholesterol levels. This association was slightly attenuated after adjustment for maternal BMI and, to a greater extent, after accounting for child lifestyle factors. Given this cluster primarily reflected total rather than phosphorylated protein levels, it may capture baseline signaling capacity or shared maternal-child metabolic milieu rather than an independent contributor to offspring cardiometabolic health. Strengths and Limitations A major strength of this study is the use of absolute quantification for a panel of placental proteins involved in key signaling and nutrient transport pathways; and the measurement of protein phosphorylation at specific serine and tyrosine residues that indicate activation. This approach allowed for direct comparison of expression levels and pathway activation across proteins and participants, enhancing biological interpretability. Another strength is the focus on a generally healthy pregnancy cohort, with a low prevalence of GDM (n=5) and maternal obesity (n=13). This low-risk context enabled identification of protein clustering patterns that likely reflect physiological coordination rather than pathologic dysregulation. This focus is especially important given that many canonical signaling relationships observed in other metabolic tissues—such as liver, adipose, or skeletal muscle—do not always translate directly to the placenta. By analyzing protein clusters in a healthy cohort, we were able to examine signaling patterns that may reflect normal physiological coordination of placental function, an area often underexplored in favor of pathophysiologic models. This perspective is particularly valuable in light of the limited foundational knowledge about normal signaling regulation in the human placenta.(67) However, these strengths must be considered alongside limitations. The clustering patterns and their associations with offspring outcomes observed in this study may not generalize to populations with greater metabolic or obstetric risk. Future work is needed to examine whether these placental signaling profiles are preserved, amplified, or disrupted in pregnancies complicated by conditions such as obesity, diabetes, or hypertension. We acknowledge that while consensus clustering provided the most robust model fit in this context, all data reduction techniques involve trade-offs and may oversimplify complex biological relationship or obscure single protein-outcome associations. Thus, the use of data reduction versus traditional single protein models should be informed by the study objectives. In addition, the relatively small sample size may have reduced the precision of effect estimates. To mitigate this, we used multiple imputation to complete missing data among those with follow-up, which maximized the sample size for all follow-up outcomes. Lastly, the study involved multiple tests of association between placental protein clusters and a range of childhood cardiometabolic outcomes. While this increases potential for false positives, we prioritized interpretation of results that were consistent across complete case and imputed models and robust to multivariable adjustment. For transparency, we present all models tested and have highlighted that findings should be considered hypothesis-generating, and none should be regarded as conclusive evidence.(68) Conclusions We identified distinct placental signaling patterns using consensus clustering, WGCNA, and PCA. Among these, latent signaling patterns derived from consensus clustering best predicted neonatal fat mass percentage. When linking the clusters of interest to offspring metabolic outcomes at median age 4.8 years, clusters that captured signaling across several pathways of insulin/growth factor signaling, stress/inflammation, and mitochondrial biogenesis were associated with fat mass, HOMA-IR, and triglycerides. These findings highlight the value of studying coordinated placental signaling in shaping early metabolic health trajectories. While these associations should be interpreted in light of the study’s modest sample size and observational design, the results lay a foundation for future mechanistic research that could inform etiological insight and early-life predictors of long-term cardiometabolic risk. Abbreviations RPS6, Ribosomal protein S6 S6K1, p70 ribosomal protein S6 kinase 1 4E-BP1, Eukaryotic translation initiation factor 4E-binding protein 1 Akt, Protein kinase B PKCα, Protein kinase C-alpha IRβ, Insulin receptor beta IGF-1r, Insulin-like growth factor 1 receptor GSK3β, Glycogen synthase kinase-3 beta ERK, Extracellular signal-regulated kinase Pro-caspase 1, Pro-caspase 1 STAT3, Signal transducer and activator of transcription 3 JNK1/2, c-Jun N-terminal kinase 1 and 2 IL-1β, Interleukin-1 beta p38 MAPK, p38 mitogen-activated protein kinase AMPK, AMP-activated protein kinase eIF2α, Eukaryotic initiation factor 2 alpha OGT, O-linked N-acetylglucosamine transferase PI3K, Phosphoinositide 3-kinase 11β-HSD1, 11β-hydroxysteroid dehydrogenase type 1 PGC-1α, Peroxisome proliferator-activated receptor gamma coactivator 1-alpha mTOR, Mechanistic Target of Rapamycin Declarations Ethics approval and consent to participate: The Healthy Start Study (ClinicalTrials.gov; NCT02273297) was approved by the Colorado Multiple Institutional Review Board at the University of Colorado Hospital. All participants gave written informed consent at enrollment. Consent for publication: Not applicable Availability of data and materials: Code can be found at https://github.com/Ecfrancis13/unsupervised-data-reduction. The datasets generated and/or analysed during the current study are not publicly available due to informed consent, but are available on reasonable request at https://healthystartstudy.org/for-investigators Competing interests: The authors declare that they have no competing interests Funding: This research was funded by the Eunice Kennedy Shriver National Institute of Child Health and Human Development grant number R00 HD108272 and the National Institute of Diabetes, Digestive, and Kidney Diseases R01 DK068001 Authors' contributions: ECF conceptualized the study, analyzed the data, and drafted the manuscript. TJ supervised the placental collection, protein assays, and provided critical review of the manuscript and revisions. DD conceptualized oversees the broader Healthy Start Study and provided critical review of the manuscript. WP, KEB, and LEG edited the manuscript, supported the interpretation of results, and provided critical review of the final version. Acknowledgements: We thank the Healthy Start participants and staff for the valuable contributions to the study. References Burton GJ, Fowden AL, Thornburg KL. Placental Origins of Chronic Disease. Physiological reviews. 2016;96(4):1509-65. Brett K, Ferraro Z, Holcik M, Adamo K. Prenatal physical activity and diet composition affect the expression of nutrient transporters and mTOR signaling molecules in the human placenta. Placenta. 2015;36(2):204-12. Thakali KM, Zhong Y, Cleves M, Andres A, Shankar K. Associations between maternal body mass index and diet composition with placental DNA methylation at term. Placenta. 2020;93:74-82. Kyriakis JM, Avruch J. Mammalian MAPK Signal Transduction Pathways Activated by Stress and Inflammation: A 10-Year Update. Physiological reviews. 2012;92(2):689-737. Wesolowski SR, Hay WW, Jr. Role of placental insufficiency and intrauterine growth restriction on the activation of fetal hepatic glucose production. Mol Cell Endocrinol. 2016;435:61-8. Singhal A. Does early growth affect long-term risk factors for cardiovascular disease? Nestle Nutr Workshop Ser Pediatr Program. 2010;65:55-64; discussion -9. Monteiro PO, Victora CG. Rapid growth in infancy and childhood and obesity in later life--a systematic review. Obes Rev. 2005;6(2):143-54. Baird J, Fisher D, Lucas P, Kleijnen J, Roberts H, Law C. Being big or growing fast: systematic review of size and growth in infancy and later obesity. BMJ. 2005;331(7522):929. Freedman DS, Dietz WH, Srinivasan SR, Berenson GS. The relation of overweight to cardiovascular risk factors among children and adolescents: the Bogalusa Heart Study. Pediatrics. 1999;103(6 Pt 1):1175-82. Freedman DS, Khan LK, Serdula MK, Dietz WH, Srinivasan SR, Berenson GS. The relation of childhood BMI to adult adiposity: the Bogalusa Heart Study. Pediatrics. 2005;115(1):22-7. Juhola J, Magnussen CG, Viikari JSA, Kähönen M, Hutri-Kähönen N, Jula A, et al. Tracking of Serum Lipid Levels, Blood Pressure, and Body Mass Index from Childhood to Adulthood: The Cardiovascular Risk in Young Finns Study. The Journal of pediatrics. 2011;159(4):584-90. Zinner SH, Rosner B, Oh W, Kass EH. Significance of blood pressure in infancy. Familial aggregation and predictive effect on later blood pressure. Hypertension. 1985;7(3 Pt 1):411-6. Ward ZJ, Long MW, Resch SC, Giles CM, Cradock AL, Gortmaker SL. Simulation of Growth Trajectories of Childhood Obesity into Adulthood. New England Journal of Medicine. 2017;377(22):2145-53. Ibáñez L, Ong K, Dunger DB, de Zegher F. Early development of adiposity and insulin resistance after catch-up weight gain in small-for-gestational-age children. The Journal of clinical endocrinology and metabolism. 2006;91(6):2153-8. Juhola J, Magnussen CG, Viikari JS, Kahonen M, Hutri-Kahonen N, Jula A, et al. Tracking of serum lipid levels, blood pressure, and body mass index from childhood to adulthood: the Cardiovascular Risk in Young Finns Study. J Pediatr. 2011;159(4):584-90. Hampl SE, Hassink SG, Skinner AC, Armstrong SC, Barlow SE, Bolling CF, et al. Clinical Practice Guideline for the Evaluation and Treatment of Children and Adolescents With Obesity. Pediatrics. 2023;151(2). Chen L, Guilmette J, Luo ZC, Cloutier A, Wang WJ, Yang MN, et al. Placental 11beta-HSD2 and Cardiometabolic Health Indicators in Infancy. Diabetes Care. 2019;42(5):964-71. Keleher MR, Erickson K, Kechris K, Yang IV, Dabelea D, Friedman JE, et al. Associations between the activity of placental nutrient-sensing pathways and neonatal and postnatal metabolic health: the ECHO Healthy Start cohort. Int J Obes (Lond). 2020. Keleher MR, Erickson K, Smith HA, Kechris KJ, Yang IV, Dabelea D, et al. Placental Insulin/IGF-1 Signaling, PGC-1alpha, and Inflammatory Pathways Are Associated With Metabolic Outcomes at 4-6 Years of Age: The ECHO Healthy Start Cohort. Diabetes. 2021;70(3):745-51. Gardebjer EM, Cuffe JS, Pantaleon M, Wlodek ME, Moritz KM. Periconceptional alcohol consumption causes fetal growth restriction and increases glycogen accumulation in the late gestation rat placenta. Placenta. 2014;35(1):50-7. Sferruzzi-Perri AN, Sandovici I, Constancia M, Fowden AL. Placental phenotype and the insulin-like growth factors: resource allocation to fetal growth. J Physiol. 2017;595(15):5057-93. Rosario FJ, Jansson N, Kanai Y, Prasad PD, Powell TL, Jansson T. Maternal Protein Restriction in the Rat Inhibits Placental Insulin, mTOR, and STAT3 Signaling and Down-Regulates Placental Amino Acid Transporters. Endocrinology. 2011;152(3):1119-29. Jansson T, Powell TL. Human Placental Transport in Altered Fetal Growth: Does the Placenta Function as a Nutrient Sensor? – A Review. Placenta. 2006;27:91-7. Pantham P, Rosario FJ, Weintraub ST, Nathanielsz PW, Powell TL, Li C, et al. Down-Regulation of Placental Transport of Amino Acids Precedes the Development of Intrauterine Growth Restriction in Maternal Nutrient Restricted Baboons. Biology of reproduction. 2016;95(5):98. Dumolt JH, Powell TL, Jansson T. Placental Function and the Development of Fetal Overgrowth and Fetal Growth Restriction. Obstet Gynecol Clin North Am. 2021;48(2):247-66. Calejman CM, Doxsey WG, Fazakerley DJ, Guertin DA. Integrating adipocyte insulin signaling and metabolism in the multi-omics era. Trends in Biochemical Sciences. 2022;47(6):531-46. Paquette AG, Hood L, Price ND, Sadovsky Y. Deep phenotyping during pregnancy for predictive and preventive medicine. Science Translational Medicine. 2020;12(527):eaay1059. Wilkerson MD, Hayes DN. ConsensusClusterPlus: a class discovery tool with confidence assessments and item tracking. Bioinformatics. 2010;26(12):1572-3. Langfelder P, Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics. 2008;9(1):559. Harrod CS, Fingerlin TE, Chasan-Taber L, Reynolds RM, Glueck DH, Dabelea D. Exposure to prenatal smoking and early-life body composition: the healthy start study. Obesity (Silver Spring). 2015;23(1):234-41. Keleher MR, Erickson K, Kechris K, Yang IV, Dabelea D, Friedman JE, et al. Associations between the activity of placental nutrient-sensing pathways and neonatal and postnatal metabolic health: the ECHO Healthy Start cohort. International Journal of Obesity. 2020. Starling AP, Brinton JT, Glueck DH, Shapiro AL, Harrod CS, Lynch AM, et al. Associations of maternal BMI and gestational weight gain with neonatal adiposity in the Healthy Start study. The American Journal of Clinical Nutrition. 2015;101(2):302-9. Francis EC, Dabelea D, Shankar K, Perng W. Maternal diet quality during pregnancy is associated with biomarkers of metabolic risk among male offspring. Diabetologia. 2021. Institute of Medicine. Weight gain during pregnancy: reexamining the guidelines. Washington, DC; 2009. Subar AF, Kirkpatrick SI, Mittl B, Zimmerman TP, Thompson FE, Bingley C, et al. The Automated Self-Administered 24-hour dietary recall (ASA24): a resource for researchers, clinicians, and educators from the National Cancer Institute. J Acad Nutr Diet. 2012;112(8):1134-7. Shapiro AL, Kaar JL, Crume TL, Starling AP, Siega-Riz AM, Ringham BM, et al. Maternal diet quality in pregnancy and neonatal adiposity: the Healthy Start Study. International journal of obesity (2005). 2016;40(7):1056-62. West KA, Schmid R, Gauglitz JM, Wang M, Dorrestein PC. foodMASST a mass spectrometry search tool for foods and beverages. npj Science of Food. 2022;6(1):22. Francis EC, Dabelea D, Boyle KE, Jansson T, Perng W. Maternal Diet Quality Is Associated with Placental Proteins in the Placental Insulin/Growth Factor, Environmental Stress, Inflammation, and mTOR Signaling Pathways: The Healthy Start ECHO Cohort. J Nutr. 2022;152(3):816-25. Guenther PM, Kirkpatrick SI, Reedy J, Krebs-Smith SM, Buckman DW, Dodd KW, et al. The Healthy Eating Index-2010 is a valid and reliable measure of diet quality according to the 2010 Dietary Guidelines for Americans. The Journal of nutrition. 2014;144(3):399-407. Guenther PM, Casavale KO, Reedy J, Kirkpatrick SI, Hiza HA, Kuczynski KJ, et al. Update of the Healthy Eating Index: HEI-2010. J Acad Nutr Diet. 2013;113(4):569-80. Fields DA, Allison DB. Air-Displacement Plethysmography Pediatric Option in 2–6 Years Old Using the Four-Compartment Model as a Criterion Method. Obesity. 2012;20(8):1732-7. World Health Organizaton. WHO child growth standards: height-for-age, weight-for-age, weight-for-length, weight-for-height and body mass index for-age: methods and development. Geneva, Switzerland: World Health Organizaton; 2006. de Onis M, Garza C, Victora C, Bhan M, Norum K. The WHO Multicentre Growth Reference Study (MGRS): rationale, planning, and implementaton. Food Nutr Bull. 2004;25(Suppl1):S1-S89. Hediger ML, Overpeck MD, Ruan WJ, Troendle JF. Early infant feeding and growth status of US-born infants and children aged 4-71 mo: analyses from the third National Health and Nutrition Examination Survey, 1988-1994. The American journal of clinical nutrition. 2000;72(1):159-67. Rowlands AV. Accelerometer assessment of physical activity in children: an update. Pediatr Exerc Sci. 2007;19(3):252-66. van Cauwenberghe E, Labarque V, Trost SG, de Bourdeaudhuij I, Cardon G. Calibration and comparison of accelerometer cut points in preschool children. International journal of pediatric obesity : IJPO : an official journal of the International Association for the Study of Obesity. 2011;6(2-2):e582-e9. Steinberger J, Daniels SR, Hagberg N, Isasi CR, Kelly AS, Lloyd-Jones D, et al. Cardiovascular Health Promotion in Children: Challenges and Opportunities for 2020 and Beyond: A Scientific Statement From the American Heart Association. Circulation. 2016;134(12):e236-55. Oken E, Kleinman KP, Rich-Edwards J, Gillman MW. A nearly continuous measure of birth weight for gestational age using a United States national reference. BMC Pediatrics. 2003;3(1):6. WHO Multicentre Growth Reference Study Group. WHO Child Growth Standards: Growth velocity based on weight, length and head circumference: Methods and development. Geneva: World Health Organization; 2009. Zwick RW, Velicer FW. Comparison of five rules for determining the number of components to retain. 1986;99:432-42. Lazo-de-la-Vega-Monroy M-L, Mata-Tapia K-A, Garcia-Santillan J-A, Corona-Figueroa M-A, Gonzalez-Dominguez M-I, Gomez-Zapata H-M, et al. Association of placental nutrient sensing pathways with birth weight. Reproduction. 2020;160(3):455-68. Jansson N, Rosario FJ, Gaccioli F, Lager S, Jones HN, Roos S, et al. Activation of Placental mTOR Signaling and Amino Acid Transporters in Obese Women Giving Birth to Large Babies. The Journal of Clinical Endocrinology & Metabolism. 2013;98(1):105-13. Zhang BB, Zhou G, Li C. AMPK: An Emerging Drug Target for Diabetes and the Metabolic Syndrome. Cell Metabolism. 2009;9(5):407-16. Dimasuay KG, Boeuf P, Powell TL, Jansson T. Placental Responses to Changes in the Maternal Environment Determine Fetal Growth. Frontiers in physiology. 2016;7:12-. Ma Y, Zhu MJ, Uthlaut AB, Nijland MJ, Nathanielsz PW, Hess BW, et al. Upregulation of growth signaling and nutrient transporters in cotyledons of early to mid-gestational nutrient restricted ewes. Placenta. 2011;32(3):255-63. Gaccioli F, White V, Capobianco E, Powell TL, Jawerbaum A, Jansson T. Maternal Overweight Induced by a Diet with High Content of Saturated Fat Activates Placental mTOR and eIF2alpha Signaling and Increases Fetal Growth in Rats. Biology of reproduction. 2013;89(4). Kola B, Grossman AB, Korbonits M, Korbonits M. The Role of AMP-Activated Protein Kinase in Obesity. Frontiers of hormone research. 2008;36:198-211. Kavitha JV, Rosario FJ, Nijland MJ, McDonald TJ, Wu G, Kanai Y, et al. Down-regulation of placental mTOR, insulin/IGF-I signaling, and nutrient transporters in response to maternal nutrient restriction in the baboon. FASEB journal : official publication of the Federation of American Societies for Experimental Biology. 2014;28(3):1294-305. Tsai K, Tullis B, Jensen T, Graff T, Reynolds P, Arroyo J. Differential expression of mTOR related molecules in the placenta from gestational diabetes mellitus (GDM), intrauterine growth restriction (IUGR) and preeclampsia patients. Reproductive Biology. 2021;21(2):100503. Kramer AC, Jansson T, Bale TL, Powell TL. Maternal-fetal cross-talk via the placenta: influence on offspring development and metabolism. Development. 2023;150(20). Lyons A, Coleman M, Riis S, Favre C, O'Flanagan CH, Zhdanov AV, et al. Insulin-like growth factor 1 signaling is essential for mitochondrial biogenesis and mitophagy in cancer cells. The Journal of biological chemistry. 2017;292(41):16983-98. Zhang Y, Castellani LW, Sinal CJ, Gonzalez FJ, Edwards PA. Peroxisome proliferator-activated receptor-gamma coactivator 1alpha (PGC-1alpha) regulates triglyceride metabolism by activation of the nuclear receptor FXR. Genes Dev. 2004;18(2):157-69. Liang H, Ward WF. PGC-1alpha: a key regulator of energy metabolism. Adv Physiol Educ. 2006;30(4):145-51. Cheong JN, Wlodek ME, Moritz KM, Cuffe JS. Programming of maternal and offspring disease: impact of growth restriction, fetal sex and transmission across generations. J Physiol. 2016;594(17):4727-40. Petersen MC, Shulman GI. Mechanisms of insulin action and insulin resistance. Physiological reviews. 2018. Le TKC, Dao XD, Nguyen DV, Luu DH, Bui TMH, Le TH, et al. Insulin signaling and its application. Frontiers in Endocrinology. 2023;Volume 14 - 2023. Herrera CL, Kim MJ, Do QN, Owen DM, Fei B, Twickler DM, et al. The human placenta project: Funded studies, imaging technologies, and future directions. Placenta. 2023;142:27-35. Rothman KJ. No adjustments are needed for multiple comparisons. Epidemiology (Cambridge, Mass). 1990;1(1):43-6. van den Berg RA, Hoefsloot HCJ, Westerhuis JA, Smilde AK, van der Werf MJ. Centering, scaling, and transformations: improving the biological information content of metabolomics data. BMC Genomics. 2006;7(1):142. Revelle W. psych: Procedures for Psychological, Psychometric, and Personality Research. Evanston, IL: Northwestern University; 2025. Haggarty P. Placental regulation of fatty acid delivery and its effect on fetal growth--a review. Placenta. 2002;23 Suppl A:S28-38. Wagenmakers EJ, Farrell S. AIC model selection using Akaike weights. Psychon Bull Rev. 2004;11(1):192-6. Rubin DB. Multiple imputation for nonresponse in surveys. New York ;: Wiley; 1987. xxix, 258 p. p. Additional Declarations No competing interests reported. Supplementary Files SupplementalMaterialDataReductionPlacentaandChildMetabolicHealthHSII10222025.docx Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 08 Dec, 2025 Reviews received at journal 07 Dec, 2025 Reviews received at journal 02 Dec, 2025 Reviews received at journal 25 Nov, 2025 Reviewers agreed at journal 23 Nov, 2025 Reviewers agreed at journal 21 Nov, 2025 Reviewers agreed at journal 11 Nov, 2025 Reviewers invited by journal 04 Nov, 2025 Editor assigned by journal 23 Oct, 2025 Submission checks completed at journal 23 Oct, 2025 First submitted to journal 22 Oct, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7926327","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":543604648,"identity":"258f781b-3d0b-42a3-aee0-51f9b06edd57","order_by":0,"name":"Ellen C. Francis","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABCklEQVRIiWNgGAWjYBAC+3bmBiD1H8w5wFDxDy7D2IBDiwEzWOoAVMuZwww8JGlhYGwjTkvj4wKGA/n8s5sfHvw573Difv7Fxz7+YLCR3XAAh1+YGZuNZzD8sZxx55jBYd5t/xJ7JJ4lz+ZhSDPGpQVoS5s0D8MBA4YbOQyHGbf9AGo5Y8zMwHA4EY+W9t8gLfJALQd/zjkD1HL+M+MPhv/4tLQxg7QYALUc4G04ltjD38MMDIMD+LQ0S/MYHDAwvJFmcJjn2GHjnhtsxsw8BsnGM3F5v7354GeeigMGcjeSH3/8UXNYtr3/8GPGHxV2sn04tEDtQuZIJKCLEAT8eE0fBaNgFIyCEQgAIotkqm21DlYAAAAASUVORK5CYII=","orcid":"","institution":"Rutgers, The State University of New Jersey","correspondingAuthor":true,"prefix":"","firstName":"Ellen","middleName":"C.","lastName":"Francis","suffix":""},{"id":543604649,"identity":"fe26a0b6-f945-4dc9-94fc-c1497bb84590","order_by":1,"name":"Kristen E. Boyle","email":"","orcid":"","institution":"Lifecourse Epidemiology of Adiposity and Diabetes Center","correspondingAuthor":false,"prefix":"","firstName":"Kristen","middleName":"E.","lastName":"Boyle","suffix":""},{"id":543604650,"identity":"4d9ff54f-9722-4da6-9e37-8882bdffe058","order_by":2,"name":"Lauren E. Gyllenhammer","email":"","orcid":"","institution":"University of California, Irvine","correspondingAuthor":false,"prefix":"","firstName":"Lauren","middleName":"E.","lastName":"Gyllenhammer","suffix":""},{"id":543604654,"identity":"24931d0e-9ef5-4cce-8228-0a416f8d614d","order_by":3,"name":"Dana Dabelea","email":"","orcid":"","institution":"Lifecourse Epidemiology of Adiposity and Diabetes Center","correspondingAuthor":false,"prefix":"","firstName":"Dana","middleName":"","lastName":"Dabelea","suffix":""},{"id":543604655,"identity":"300df117-339f-4407-a8d0-22ec6692999e","order_by":4,"name":"Thomas Jansson","email":"","orcid":"","institution":"Department of Obstetrics and Gynecology","correspondingAuthor":false,"prefix":"","firstName":"Thomas","middleName":"","lastName":"Jansson","suffix":""},{"id":543604656,"identity":"e1c81519-2e6e-4c3f-9b8f-fe87eb4a40c4","order_by":5,"name":"Wei Perng","email":"","orcid":"","institution":"Lifecourse Epidemiology of Adiposity and Diabetes Center","correspondingAuthor":false,"prefix":"","firstName":"Wei","middleName":"","lastName":"Perng","suffix":""}],"badges":[],"createdAt":"2025-10-22 19:38:18","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7926327/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7926327/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":96244702,"identity":"295e2153-27b9-42bb-9cca-6169aa695204","added_by":"auto","created_at":"2025-11-19 07:19:06","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":447937,"visible":true,"origin":"","legend":"","description":"","filename":"MSDataReductionPlacentaandChildMetabolicHealthHSII10222025.docx","url":"https://assets-eu.researchsquare.com/files/rs-7926327/v1/af3085d539d105d83ec3521d.docx"},{"id":96243960,"identity":"471bc176-4616-4d27-b4fa-808aeeae00d9","added_by":"auto","created_at":"2025-11-19 07:17:24","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8997,"visible":true,"origin":"","legend":"","description":"","filename":"536b21b5841f4c7f8acb48860772a3bb.json","url":"https://assets-eu.researchsquare.com/files/rs-7926327/v1/cf7761ee066946f480913a5b.json"},{"id":96244088,"identity":"19a2074a-200f-410f-8856-631e827302f5","added_by":"auto","created_at":"2025-11-19 07:17:41","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":58953,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementalMaterialDataReductionPlacentaandChildMetabolicHealthHSII10222025.docx","url":"https://assets-eu.researchsquare.com/files/rs-7926327/v1/e8dac26a736d494f85498a08.docx"},{"id":95939250,"identity":"33096d28-635d-4f3f-a47e-1dd6f00784cc","added_by":"auto","created_at":"2025-11-14 16:07:07","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":216006,"visible":true,"origin":"","legend":"","description":"","filename":"536b21b5841f4c7f8acb48860772a3bb1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7926327/v1/6c880ba50207c30dc90ce4ec.xml"},{"id":96244483,"identity":"570d8ed2-04ed-4dde-a046-3067f7f4447e","added_by":"auto","created_at":"2025-11-19 07:18:35","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":223894,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7926327/v1/8d4a3224a09a0e22d8f6c929.jpeg"},{"id":95939243,"identity":"9975c63f-d35a-416a-bfc6-68561e825b15","added_by":"auto","created_at":"2025-11-14 16:07:07","extension":"png","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29423,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7926327/v1/fe28fedd76c28517b6cc6d79.png"},{"id":95939238,"identity":"39d151f6-b830-48ef-a6b7-7b6148a9a9ba","added_by":"auto","created_at":"2025-11-14 16:07:07","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":9034,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7926327/v1/c7cde3ebea4176a59d33f623.png"},{"id":96244407,"identity":"ee84a1d9-eb7d-486b-9e68-d324e52063fb","added_by":"auto","created_at":"2025-11-19 07:18:23","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":14953,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7926327/v1/35e02553e6656900f6001041.png"},{"id":95939247,"identity":"06962f2f-4bd0-4f6c-b217-04282bd321ac","added_by":"auto","created_at":"2025-11-14 16:07:07","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":38502,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7926327/v1/4e83061b1d819356bfb060ab.png"},{"id":95939245,"identity":"73ed8f70-cede-4b4b-b157-34de88919e41","added_by":"auto","created_at":"2025-11-14 16:07:07","extension":"png","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":14627,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7926327/v1/96a7afcbeb6ab14dfa10a3ee.png"},{"id":96243281,"identity":"c94ef070-8236-4022-9fd5-ddfb14487074","added_by":"auto","created_at":"2025-11-19 07:15:59","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8730,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7926327/v1/7a39745b4da497ed603864fb.png"},{"id":95939249,"identity":"2d473ad5-fdc2-465a-8bd4-9e3b1b829b97","added_by":"auto","created_at":"2025-11-14 16:07:07","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":7652,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-7926327/v1/cea03af54290033a1d5940b7.png"},{"id":95939251,"identity":"940682a0-edfe-45c2-bc49-9355c824c447","added_by":"auto","created_at":"2025-11-14 16:07:07","extension":"xml","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":213149,"visible":true,"origin":"","legend":"","description":"","filename":"536b21b5841f4c7f8acb48860772a3bb1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7926327/v1/8a7ef34c7e5e3bd54ca78ce2.xml"},{"id":95939252,"identity":"9881c7b7-024d-4504-90e6-4e94d5f15008","added_by":"auto","created_at":"2025-11-14 16:07:07","extension":"html","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":223949,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7926327/v1/a02125f410c491d37ad12beb.html"},{"id":96243107,"identity":"6c6ec7ff-df66-4fb2-bbe6-b15583afc3ad","added_by":"auto","created_at":"2025-11-19 07:15:32","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":186951,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eAnalytic workflow for evaluating placental signaling patterns and their association with early childhood cardiometabolic health.\u003cbr\u003e\nThe analysis proceeded in five sequential steps.\u003cbr\u003e\n \u003c/strong\u003e(1) Placental protein data were preprocessed by normalizing expression levels to total placental protein and calculating phospho-to-total protein ratios to capture pathway activation. Missing values were imputed using multiple imputation with chained equations (MICE) based on classification and regression trees.\u003cbr\u003e\n(2) To derive latent variables summarizing placental signaling, we applied three unsupervised data reduction methods—consensus clustering, weighted gene co-expression network analysis (WGCNA), and principal component analysis (PCA)—each generating summarized scores from protein expression profiles.\u003cbr\u003e\n(3) To evaluate model performance, we tested associations between the reduced dimensions and offspring fat mass percentage (neonatal and early childhood) using multivariable linear regression adjusted for child sex and age. Model fit was assessed using Akaike information criterion (AIC), adjusted R², root mean square error (RMSE), paired t-tests with bootstrap resampling, and Akaike weights.\u003cbr\u003e\n(4) The best-performing method was selected based on its ability to explain variation in offspring fat mass while balancing fit and parsimony.\u003cbr\u003e\n(5) Summary scores from the selected data reduction method were used in adjusted models to examine associations with early childhood fat mass and cardiometabolic outcomes. All models were evaluated using both complete-case and multiply imputed data\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7926327/v1/172261dde96b682a292b9995.png"},{"id":95939241,"identity":"065a91f8-de45-4844-ae09-f68353461128","added_by":"auto","created_at":"2025-11-14 16:07:07","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":292184,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eVisualization of placenta protein assignments to clusters, modules, or components based on 33 placental proteins and phosphorylated-to-total protein ratio.\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eA.\u003c/strong\u003e \u003cstrong\u003eConsensus matrix heatmap from consensus clustering of placental proteins (k = 4). \u003c/strong\u003eConsensus clustering was perfomred on scaled protein data. Each cell in the matrix represents the proportion of resampling iterations where both proteins were assigned to the same cluster across 1,000 resampling iterations using hierarchical clustering with Spearman distance. Darker blue indicates higher co-clustering stability across subsamples. The matrix is ordered by clustering assignment. This analysis identified four stable clusters of proteins.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eB.\u003c/strong\u003e \u003cstrong\u003eWeighted gene correlation network analysis (WGCNA) dendrogram of placental proteins (modules = 5). \u003c/strong\u003eHierarchical clustering dendrogram of scaled protein data constructed using a signed network based on Spearman correlation and topological overlap. The soft-thresholding power was set to 6 to approximate scale-free topology, and a minimum module size of 3 proteins was used based on relatively small input protein dataset of 34. Modules (clusters of highly interconnected proteins) were identified using dynamic tree cutting and merged at a cut height of 0.25. Color bars beneath the dendrogram indicate final module assignments after merging. This network-based approach groups proteins with similar expression patterns and network connectivity.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eC.\u003c/strong\u003e \u003cstrong\u003eHeatmap of protein loadings on principal components (retained componenets = 9). \u003c/strong\u003ePrincipal component analysis (PCA) was performed on log-transformed and scaled protein data to identify components capturing shared variance across 33 normalized total and ratio- proteins. Parallel anaysis and Kaiser’s rule were used to select the number of components to retain for analysis (PC=9). This heatmap displays the loadings (correlations between proteins and PCs) for the top 18 proteins with the highest cumulative absolute loadings across PCs 1–9. Warm colors (red) indicate strong positive contributions, and cool colors (blue) indicate strong negative contributions to each principal component. Each component reflects a distinct axis of shared variation.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7926327/v1/a458bd7fc390aa8d3934d37f.png"},{"id":96363109,"identity":"cb2845e3-f55d-4798-bb1e-6c60dc99d0fb","added_by":"auto","created_at":"2025-11-20 10:04:35","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2342297,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7926327/v1/8798c152-a9e8-4c2a-b9a9-791435ee002c.pdf"},{"id":95939236,"identity":"4365aae8-14b4-44a4-a21a-f23c416b45f9","added_by":"auto","created_at":"2025-11-14 16:07:07","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":58953,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementalMaterialDataReductionPlacentaandChildMetabolicHealthHSII10222025.docx","url":"https://assets-eu.researchsquare.com/files/rs-7926327/v1/f0e720fd7807ac24a8b37262.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Unsupervised Approaches to Placental Protein Clustering: Which Best Captures Signals Linked to Childhood Metabolic Health?","fulltext":[{"header":"Background","content":"\u003cp\u003eThe activation of placental signaling pathways that regulate nutrient transport is central to fetal development and may have lasting implications for metabolic health across the life course.(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) Animal studies have demonstrated that disruptions in placental signaling pathways can impair fetal growth. For example, increased nutrient transport via mTOR activation has been associated with fetal overgrowth, while greater placental stress activation (e.g., JNK) has been observed in intrauterine growth restriction\u0026mdash;a condition linked to increased fat mass and a less favorable cardiometabolic profile later in life.(\u003cspan additionalcitationids=\"CR3 CR4\" citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e)\u003c/p\u003e\u003cp\u003eImportantly, poor cardiometabolic traits in childhood track across the life course and are predictors of future obesity,(\u003cspan additionalcitationids=\"CR7 CR8 CR9\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e) type 2 diabetes, and metabolic syndrome,(\u003cspan additionalcitationids=\"CR12 CR13 CR14\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e) highlighting the importance of early identification of biological risk before overt complications arise.(\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e) In human pregnancies, variation in the above placental signaling pathways have also been associated with neonatal and childhood metabolic outcomes, including adiposity, lipid profiles, and insulin sensitivity.(\u003cspan additionalcitationids=\"CR18 CR19 CR20 CR21 CR22 CR23\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e) However, much of this work has focused on discrete proteins or single pathways, often in the context of maternal complications such as obesity or gestational diabetes (GDM).(\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e) As a result, there is limited understanding of how the interrelationship among these pathways in healthy, low-risk pregnancies\u0026mdash;where subtle signaling variation may still be biologically meaningful. As biological pathways operate in concert, understanding their coordination may be particularly important in normal low-risk pregnancies where severe dysregulation of a single pathway is likely to be absent. These subtle shifts may nonetheless influence long-term health trajectories, given that gestation is a sensitive period for developmental programming.\u003c/p\u003e\u003cp\u003eAdvances in omics science have enabled the systematic measurement of multiple biological pathways concurrently. In parallel, analytic frameworks such as clustering, network analyses, and data reduction have been developed to identify patterns of co-expression or co-regulation within high-dimensional datasets. These unsupervised approaches can reveal latent structure across interacting pathways, offering a systems-level view of molecular signaling that complements traditional single-protein models.(\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e) Despite their utility for summarizing large amounts of data, we are not aware of studies that have applied or compared such methods to classically measured protein data, such as those obtained via Western assays.\u003c/p\u003e\u003cp\u003eOur primary objective was to compare three commonly used unsupervised analytic methods\u0026mdash; consensus clustering(\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e), weighted gene correlation network analysis (WGCNA)(\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e), and principal component analysis (PCA)\u0026mdash;for identifying placental protein signaling patterns potentially relevant to fetal and childhood metabolic outcomes. We applied these methods to a panel of placental proteins quantified using Simple Western Assays (WES) and evaluated their performance in models predicting neonatal and early childhood adiposity. Secondary analyses examined associations with broader childhood cardiometabolic health outcomes.\u003c/p\u003e\u003cp\u003eThese three methods were compared because PCA one of the most used approaches for data reduction, WGCNA is a commonly used network-based technique, and consensus clustering includes internal resampling to identify cluster stability, which is not a standard step in PCA or WGCNA. These methods summarize patterns of co-expression or shared variance across multiple proteins, capturing broader signaling relationships across multiple pathways.\u003c/p\u003e"},{"header":"Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003eParticipants and sample collection\u003c/h2\u003e\u003cp\u003eWe used data from the Healthy Start Study (ClinicalTrials.gov; NCT02273297). The Healthy Start pre-birth cohort study enrolled 1,410 women aged\u0026thinsp;\u0026ge;\u0026thinsp;16 years of age at \u0026lt;\u0026thinsp;24 weeks of gestation from prenatal clinics at the University of Colorado Hospital between 2009\u0026ndash;2014. We excluded women with prior diabetes, a history of prior preterm birth\u0026thinsp;\u0026lt;\u0026thinsp;25 gestational weeks or fetal death, asthma with active steroid management, serious psychiatric illness or multiple gestation.(\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e) As part of the ancillary Healthy Start BabyBUMP Project we collected and snap froze trophoblast villi samples from placentas after delivery in a convenience subsample (N\u0026thinsp;=\u0026thinsp;111) of the Healthy Start cohort. Of these women, we excluded those women with a pre-term delivery (N\u0026thinsp;=\u0026thinsp;2) and who were completely missing offspring outcomes (N\u0026thinsp;=\u0026thinsp;1). The final sample included 108 placentas.\u003c/p\u003e\u003cp\u003eOf the 108 mother-offspring pairs, all were invited to complete a follow-up research visit in early childhood (mean offspring age 4.8\u0026thinsp;\u0026plusmn;\u0026thinsp;0.6), and 67% (n\u0026thinsp;=\u0026thinsp;72) of those children returned for an in-person visit during which weight and height were measured (details in proceeding sections). Among the children who returned for the early childhood visit, 72% (n\u0026thinsp;=\u0026thinsp;52) also completed a blood collection.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003ePlacental samples and signaling\u003c/h3\u003e\n\u003cp\u003eThe placental proteins were selected \u003cem\u003ea priori\u003c/em\u003e based on their roles in insulin/growth factor signaling, inflammation, nutrient transport, and energy sensing.(\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e) We homogenized\u0026thinsp;~\u0026thinsp;20mg of frozen trophoblast placental villus tissue in 75\u0026micro;L ice-cold buffer D (250mM sucrose, 10mM HEPES, pH 7.4) that had a 1:100 dilution of protease and phosphatase inhibitors. Next, we used Simple Western Assays with WES (ProteinSimple, Santa Clara, CA) to measure the phosphorylation and total abundance of the following proteins: ribosomal protein S6 (RPS6), p70 ribosomal protein S6 kinase 1 (S6K1), eukaryotic translation initiation factor 4E-binding protein 1 (4E-BP1), protein kinase B (Akt), protein kinase C-α (PKCα), insulin receptor β (IRβ), insulin-like growth factor 1 receptor (IGF-1r), glycogen synthase kinase-3 beta (GSK3β), extracellular signal-regulated kinase (ERK), 1/2 pro-caspase 1, signal transducer and activator of transcription 3 (STAT3), c-Jun N-terminal kinase 1 and 2 (JNK1 and JNK2), interleukin-1 beta (IL-1β), p38 mitogen-activated protein kinase (p38 MAPK), AMP-activated protein kinase (AMPK), eukaryotic initiation factor 2 alpha (eIF2α), O-linked N-acetylglucosamine transferase (OGT), phosphoinositide 3-kinase (PI3K), 11β-hydroxysteroid dehydrogenase type 1 (11β-HSD1), and peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α). We used a 0.1 mg/mL total protein concentration and ran the WES plates according to the manufacturer\u0026rsquo;s instructions, with slight modification (200V, 55m separation time). We included an equalizer sample on each plate with clean median values for each protein to control for batch variation. We also multiplexed a loading control (vinculin or β-actin) in each capillary and normalized the protein levels to it.\u003c/p\u003e\n\u003ch3\u003eMaternal and child data\u003c/h3\u003e\n\u003cdiv id=\"Sec6\" class=\"Section2\"\u003e\u003ch2\u003eMaternal lifestyle and clinical data\u003c/h2\u003e\u003cp\u003eMaternal race and ethnicity, educational attainment, parity, and smoking status during pregnancy were self-reported via questionnaire. GDM status was abstracted from medical records and all data collection are reported in detail elsewhere.(\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e) We calculated maternal pre-pregnancy BMI based on pre-pregnancy weight obtained from medical records (89%) or self-report (11%), and measured height at the first research visit. Gestational weight gain (GWG) was estimated by subtracting pre-pregnancy weight from the last clinically measured weight during pregnancy and categorized based on the Institute of Medicine 2009 Guidelines.(\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e) Beginning in the 1st trimester, maternal diet was assessed via the Automated Self-Administered 24-h Dietary Recall (ASA24).(\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e) We calculated the Healthy Eating Index-2010 (HEI)(\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e, \u003cspan additionalcitationids=\"CR37\" citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e) to capture diet quality in accordance with the 2010 Dietary Guidelines for Americans. (\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e)\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eChild metabolic outcomes\u003c/h3\u003e\n\u003cp\u003eOffspring fat mass and fat free mass at birth and in early childhood were measured using whole body air displacement plethysmography (ADP; PeaPod and BodPod, Life Measurement, Inc.) with the Pediatric Option. (\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e) Measurements for each participant were taken in triplicate and the average of the two closest measures was used for analyses. The children\u0026rsquo;s height and weight were measured by trained nurses. Age-specific body mass index (BMI) percentile were calculated according to the World Health Organization (WHO) growth reference (\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e, \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eFasting serum was used to measure triglycerides (TGs), total cholesterol, high-density lipoprotein (HDL), low-density lipoprotein (LDL), and glucose using manufacturer prepackaged enzymatic kits and the AU400e Chemistry Analyzer (Olympus America). Insulin was measured using a radioimmune assay, and leptin and adiponectin were measured using a Multiplex assay kit, all by Millipore Corporation.\u003c/p\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eChild lifestyle data\u003c/h2\u003e\u003cp\u003eEarly childhood dietary data was collected via two Automated Self-Administered 24-hour Dietary Assessment Tool recalls (1 weekend and 1 weekday, with caretaker proxy) (\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e). Nutrient and caloric intake were derived using Nutrition Data System for Research software package. Offspring physical activity was measured using wGT3X-BT ActiGraph accelerometers (Pensacola, FL) worn for 7 days during waking hours on the waist.(\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e) Average weekly moderate to vigorous physical activity was categorized as adequate (\u0026ge;\u0026thinsp;1 hour/day) using intensity cut points established in youth.(\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e) We derived the child\u0026rsquo;s HEI scores using the same procedure as for the mothers, except the 2015 U.S. Dietary Guidelines were used to align with timing of data collection.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eData and Statistical Analysis\u003c/h3\u003e\n\u003cp\u003eThe analytical workflow is shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. All data and statistical analysis were performed using R Statistical Software (v. 4.0.0 R Core Team 2024). Please note that full details on data and statistical analyses are in \u003cb\u003eSupplemental Methods\u003c/b\u003e. A brief overview is provided below.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003cb\u003eThe analysis proceeded in five sequential steps.\u003c/b\u003e\u003c/p\u003e\u003cp\u003e(\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e) Placental protein data were preprocessed by normalizing expression levels to total placental protein and calculating phospho-to-total protein ratios to capture pathway activation. Missing values were imputed using multiple imputation with chained equations (MICE) based on classification and regression trees.\u003c/p\u003e\u003cp\u003e(\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e) To derive latent variables summarizing placental signaling, we applied three unsupervised data reduction methods\u0026mdash;consensus clustering, weighted gene co-expression network analysis (WGCNA), and principal component analysis (PCA)\u0026mdash;each generating summarized scores from protein expression profiles.\u003c/p\u003e\u003cp\u003e(\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e) To evaluate model performance, we tested associations between the reduced dimensions and offspring fat mass percentage (neonatal and early childhood) using multivariable linear regression adjusted for child sex and age. Model fit was assessed using Akaike information criterion (AIC), adjusted R\u0026sup2;, root mean square error (RMSE), paired t-tests with bootstrap resampling, and Akaike weights.\u003c/p\u003e\u003cp\u003e(\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e) The best-performing method was selected based on its ability to explain variation in offspring fat mass while balancing fit and parsimony.\u003c/p\u003e\u003cp\u003e(\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e) Summary scores from the selected data reduction method were used in adjusted models to examine associations with early childhood fat mass and cardiometabolic outcomes. All models were evaluated using both complete-case and multiply imputed data\u003c/p\u003e\n\u003ch3\u003ePlacenta Data Preprocessing\u003c/h3\u003e\n\u003cp\u003ePlacental protein expression was normalized to total protein levels, and phosphorylated-to-total ratios were computed to capture signaling activation. Total protein expression values were also retained to assess co-expression patterns. A small number of samples with partial missingness were imputed using multiple imputation. Full preprocessing and imputation details are provided in \u003cb\u003eSupplemental Methods Section 1.\u003c/b\u003e\u003c/p\u003e\u003cdiv id=\"Sec11\" class=\"Section2\"\u003e\u003ch2\u003eData Reduction Methods\u003c/h2\u003e\u003cp\u003eThree unsupervised approaches were applied to reduce dimensionality of the placental signaling data: consensus clustering, WGCNA, and PCA. Each method generated latent variables summarizing placental protein patterns for downstream modeling. We retained dimensions based on established criteria for each method (e.g., cluster stability, module size, eigenvalue thresholds). See \u003cb\u003eSupplemental Methods Section 2\u003c/b\u003e for technical implementation details.\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec12\" class=\"Section2\"\u003e\u003ch2\u003eModel Comparison and Main Analyses\u003c/h2\u003e\u003cp\u003eTo compare the utility of each data reduction approach and to select the optimal method for downstream analysis, we examined associations of the latent structured generated by each with neonatal and early childhood fat mass percentage using multivariable linear regression adjusted for child age and sex. Model performance was evaluated using Akaike Information Criterion (AIC), adjusted R\u0026sup2;, and root mean squared error (RMSE). The best-performing method was selected based on consistent performance across these metrics (see \u003cb\u003eSupplemental Methods Section 3\u003c/b\u003e).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\u003ch2\u003eAssociation of Placental Clusters and Child Cardiometabolic Outcomes\u003c/h2\u003e\u003cp\u003eFollowing identification of the optimal dimension reduction approach, we investigated associations of the optimal method corresponding latent structure scores with offspring cardiometabolic outcomes.\u003c/p\u003e\u003cp\u003eHere, we specified three sequentially adjusted multivariable models using complete case (CC) and Multivariate Imputation by Chained Equations (MICE) (\u003cb\u003eSupplemental Methods Section 4)\u003c/b\u003e. Model 1 adjusted for offspring sex and age at follow-up. Model 2 adjusted for Model 1 and observed gestational weight gain and pre-pregnancy BMI as confounders of placenta-child outcome association that is not explained by the placental signaling pathways measured in our study (e.g., uteroplacental blood flow). Model 3 adjusted for Model 1 and child diet quality (total HEI score) and physical activity (minutes of moderate-to-vigorous activity) measured at follow-up to account for differences in children\u0026rsquo;s lifestyle (e.g., precision covariates).\u003c/p\u003e\u003cp\u003eWe used several criteria to evaluate the robustness of associations between placental signaling clusters and child metabolic outcomes. First, estimates had to demonstrate consistency across CC and MICE, as well as across the different covariate adjustment models, in terms of both magnitude and direction. Second, we considered associations statistically significant at a nominal alpha of 0.05, corresponding to a 95% confidence interval that excluded the null. Additionally, given the relatively small sample size and number of comparisons made, we supplemented interpretation of results with estimates that met statistical significance at alpha of 0.15\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec14\" class=\"Section2\"\u003e\u003ch2\u003eSensitivity analysis\u003c/h2\u003e\u003cp\u003eWe performed several sensitivity analyses in the \u003cem\u003edata reduction\u003c/em\u003e and \u003cem\u003eassociation testing\u003c/em\u003e phases. First, although we include R\u003csup\u003e2\u003c/sup\u003e in our assessment of the different data reduction techniques, we further assessed if variance explained in the outcome was due to differences in the number of dimensions for each reduction technique (consensus clustering\u0026thinsp;=\u0026thinsp;4, WGCNA\u0026thinsp;=\u0026thinsp;5, and PCA\u0026thinsp;=\u0026thinsp;9). The top three dimensions were chosen based on the 1st three PCs, the three WGCNA modules with the highest mean connectivity, and the three largest consensus clusters. Second, we specified joint models that included the linear combination of all placental consensus clustering scores with neonatal and childhood fat mass percentage as separate outcomes and evaluated the estimates for consistency with the series of individual models where each cluster was tested separately. Lastly, we compared differences in mean cluster scores by maternal characteristics to identify upstream factors associated with placental signaling clusters.\u003c/p\u003e\u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec16\" class=\"Section2\"\u003e\n \u003ch2\u003eCharacteristics\u003c/h2\u003e\n \u003cp\u003eMaternal and child characteristics are presented in Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e. Most of the pregnancies were from non-Hispanic white females who had completed at least some college. The mean pre-pregnancy BMI was 24.6\u0026thinsp;\u0026plusmn;\u0026thinsp;4.2 kg/m\u003csup\u003e2\u003c/sup\u003e and 10% of women developed GDM. The mean gestational age of offspring was at term (277\u0026thinsp;\u0026plusmn;\u0026thinsp;8.5 days), birth weight-for-gestational age percentile of 29.9 (24.5), and 51.9% were males. At the follow-up research visit, the mean age of children was 4.79\u0026thinsp;\u0026plusmn;\u0026thinsp;0.6 years, BMI percentile was 46.2\u0026thinsp;\u0026plusmn;\u0026thinsp;22.4, HEI score was 60.8\u0026thinsp;\u0026plusmn;\u0026thinsp;10.1, and minutes of vigorous/moderate activity per week was 48.5\u0026thinsp;\u0026plusmn;\u0026thinsp;25.3.\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\n \u003ctable id=\"Tab1\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eMaternal and offspring characteristics\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCharacteristics; mean (SD), n (%)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eOverall N\u0026thinsp;=\u0026thinsp;108\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eMaternal\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAge at index pregnancy (years)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e29.1 (6.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eRace-ethnicity\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNon-Hispanic White\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e67 (62.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNon-Hispanic Black\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e12 (11.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHispanic\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e23 (21.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eNon-Hispanic Other\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e6 (5.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eEducation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCollege Degree\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e26 (24.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGrad Degree\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e29 (26.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSome college\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e25 (23.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHS or less\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e28 (25.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003ePre-pregnancy BMI (kg/m\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e25.0 (5.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\" colspan=\"2\"\u003e\n \u003cp\u003eGestational diabetes mellitus\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eYes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e5 (4.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTotal HEI score\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e56.8 (13.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTotal gestational weight gain (lbs)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e13.5 (7.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eSmoked during pregnancy\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e8 (7.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eNeonatal\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eMale sex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e56 (51.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGestational age at delivery (days)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e277 (8.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eWeight-for-gestational age percentile \u003csup\u003eb\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e29.9 (24.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eChildhood\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAge (years)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e4.79 (0.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHealthy Eating Index (score)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e60.8 (10.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eVigorous/Moderate activity per week (minutes)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e48.5 (25.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBMI, (kg/m\u003csup\u003e2\u003c/sup\u003e) \u003csup\u003ec\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e15.2 (1.23)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBMI, (percentile)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e46.2 (22.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003ea Other includes, Asian, American Indian/Alaska natives, Hawaiian/Pacific Islanders\u003c/p\u003e\n \u003cp\u003eb Birthweight z-score specific to gestational age derived from U.S. natality reference (\u003cspan class=\"CitationRef\"\u003e48\u003c/span\u003e)\u003c/p\u003e\n \u003cp\u003ec BMI specific to age derived from WHO growth standards (\u003cspan class=\"CitationRef\"\u003e49\u003c/span\u003e)\u003c/p\u003e\n \u003cp\u003eHEI; Healthy Eating Index\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e\n \u003ch2\u003eData reduction and protein clustering\u003c/h2\u003e\n \u003cp\u003eThe placental protein data showed a modest correlation structure, with a median pairwise Spearman correlation of 0.13 (IQR: 0.06\u0026ndash;0.22). The results of the three unspurivsed data reduction methods, consensus clustering, WGCNA, and PCA, are shown in \u003cstrong\u003eFig.\u0026nbsp;2.\u003c/strong\u003e Across the three methods, several consistent patterns emerged. A subset of proteins involved in growth and metabolic signaling pathways, including IGF1r, AKT, and PGC1\u0026alpha;, frequently appeared together. Proteins associated with cellular stress or inflammatory signaling, such as total levels of STAT3, eIF2a, p38MAPK, and JNK also clustered together or loaded onto similar components. The grouping of these proteins were often combined with their phophorylated forms in growth and metabolic signaling such as JNK ratios and ERK1/2 ratios. In addition, activation of proteins involved in mTORC1 regulating translation, including phosphorylation of 4E-BP1 and RPS6, frequently appeared together across methods.\u003c/p\u003e\n \u003cp\u003eAcross both child adiposity outcomes, models using all available dimensions showed that consensus clustering consistently provided the best model fit (lowest AIC), the highest explanatory power (adjusted R\u0026sup2;), and the strongest model support (Akaike weights). For instance, when evaluating neonatal fat mass percent, consensus cluster metrics (AIC\u0026thinsp;=\u0026thinsp;589.5; Akaike weight\u0026thinsp;=\u0026thinsp;0.978) indicated it was more likely to be the best model compared to PCA (AIC\u0026thinsp;=\u0026thinsp;600.5; weight\u0026thinsp;=\u0026thinsp;0.004) and WGCNA (AIC\u0026thinsp;=\u0026thinsp;597.5; weight\u0026thinsp;=\u0026thinsp;0.018). Further, the RMSE was significantly lower for the consensus clustering model compared to WGCNA (p\u0026thinsp;\u0026lt;\u0026thinsp;2.2e-16), though not significantly different compared to PCA (p\u0026thinsp;=\u0026thinsp;0.18) (\u003cstrong\u003eSupplemental Table\u0026nbsp;1\u003c/strong\u003e). Given that consensus clustering demonstrated better model performance more consistently, we used consensus clusters and their summary scores as the main predictor.\u003c/p\u003e\n \u003cp\u003eThe four consensus clusters were labeled based on the proteins grouped into each cluster. \u003cstrong\u003e\u0026ldquo;\u003c/strong\u003e\u003cstrong\u003emTOR/AMPK\u003c/strong\u003e\u003cstrong\u003e\u0026rdquo;\u003c/strong\u003e (n\u0026thinsp;=\u0026thinsp;4 proteins) included the phosphorylated proteins from insulin/IGF signaling and energy-sensing pathways AMPK ratio and RPS6 ratio. \u003cstrong\u003e\u0026ldquo;\u003c/strong\u003e\u003cstrong\u003eIGF/Mitochondrial Biogensis\u003c/strong\u003e\u003cstrong\u003e\u0026rdquo;\u003c/strong\u003e (n\u0026thinsp;=\u0026thinsp;2 proteins) contained IGF1r and PGC1\u0026alpha; which are in insulin signlaing and mitochondrial biogenesis pathways. \u003cstrong\u003e\u0026ldquo;\u003c/strong\u003e\u003cstrong\u003ePlacental Insulin Coordination\u003c/strong\u003e\u003cstrong\u003e\u0026rdquo;\u003c/strong\u003e (n\u0026thinsp;=\u0026thinsp;15 proteins) which was the largest and comprised primarily of phosphorylated proteins from all pathways except mitochondrial biogenesis and cortisol metabolism. \u0026ldquo;\u003cstrong\u003eInflammation/Stress\u003c/strong\u003e\u003cstrong\u003e\u0026rdquo;\u003c/strong\u003e (n\u0026thinsp;=\u0026thinsp;12 proteins) was comprised primarily of total protein concentration markers of MAPK signaling, inflammatory signaling, and stress response, such as ERK1/2, p38MAPK, STAT3, JNK1, and IL-1\u0026beta;. Cluster scores represent the mean standardized (z-scored) expression of proteins within each cluster for each participant, such that higher scores indicate relatively higher expression or activation of proteins assigned to that cluster compared to the overall mean in our sample of participants.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e\n \u003ch2\u003eAssociations of placental protein clusters with child cardiometabolic outcomes\u003c/h2\u003e\n \u003cp\u003eSeveral consensus clusters showed consistent associations with childhood cardiometabolic outcomes across adjustments for child age and sex (Model 1), maternal pre-pregnancy BMI and gestational weight gain (Model 2; Table \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e), and child lifestyle factors (Model 3; \u003cstrong\u003eSupplemental Table\u0026nbsp;3\u003c/strong\u003e). Here we highlight patterns of association that showed similar directions of effect across model adjustment, recognizing that in a few instances, estimates varied in magnitude and precision between CC and MICE.\u003c/p\u003e\n \u003cp\u003eIn both CC and MICE models, a higher \u0026ldquo;\u003cem\u003emTOR/AMPK\u003c/em\u003e\u0026rdquo; cluster score was associated with lower childhood fat mass percentage (e.g., \u0026beta; = \u0026minus;\u0026thinsp;2.51%, 95% CI: \u0026minus;\u0026thinsp;4.44, \u0026minus;\u0026thinsp;0.58; p\u0026thinsp;=\u0026thinsp;0.01 in the MICE Model 2). This association was slightly attenuated after adjusting for concurrent lifestyle. Higher \u0026ldquo;\u003cem\u003eIGF/Mitochondrial Biogenesis\u003c/em\u003e\u0026rdquo; cluster scores were consistently associated with higher triglycerides in the CC models (e.g., \u0026beta;\u0026thinsp;=\u0026thinsp;17.90 mmol/l, 95% CI: 6.14, 29.60; p\u0026thinsp;\u0026lt;\u0026thinsp;0.001 in CC Model 2), with reduced precision and attenuation in MICE models (e.g., \u0026beta;\u0026thinsp;=\u0026thinsp;8.20 mmol/l, 95% CI: \u0026minus;\u0026thinsp;1.73, 18.13; p\u0026thinsp;=\u0026thinsp;0.11 in MICE Model 2). \u0026ldquo;\u003cem\u003ePlacental Insulin Coordination\u003c/em\u003e\u0026rdquo; scores trended with lower insulin levels, though confidence intervals crossed zero (e.g., \u0026beta; = \u0026minus;\u0026thinsp;2.63 pmol/l, 95% CI: \u0026minus;\u0026thinsp;5.63, 0.37; p\u0026thinsp;=\u0026thinsp;0.09 in MICE Model 2). Higher \u0026ldquo;\u003cem\u003eInflammation/Stress\u003c/em\u003e\u0026rdquo; scores were associated with lower LDL cholesterol in CC models, with confidence intervals excluding zero after adjusting for concurrent lifestyle (\u0026beta; = \u0026minus;\u0026thinsp;17.00 mmol/l, 95% CI: \u0026minus;\u0026thinsp;33.30, \u0026minus;\u0026thinsp;0.67; p\u0026thinsp;=\u0026thinsp;0.04 in CC Model 3). This association was attenuated and became imprecise in all MICE models.\u003c/p\u003e\n \u003cdiv class=\"gridtable\"\u003e\n \u003ctable id=\"Tab2\" border=\"1\"\u003e\n \u003ccaption language=\"En\"\u003e\n \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\n \u003cdiv class=\"CaptionContent\"\u003e\n \u003cp\u003eAssociation of Placental Signaling Clusters with Childhood Metabolic Health Indicators\u003c/p\u003e\n \u003c/div\u003e\n \u003c/caption\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\" colspan=\"6\"\u003e\n \u003cp\u003eMODEL 1\u003c/p\u003e\n \u003cp\u003e(age \u0026amp; sex)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\" colspan=\"6\"\u003e\n \u003cp\u003eMODEL 2\u003c/p\u003e\n \u003cp\u003e(Model 1\u0026thinsp;+\u0026thinsp;Maternal pre-pregnancy BMI)\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003cth align=\"left\" rowspan=\"2\"\u003e\n \u003cp\u003eCluster,\u003c/p\u003e\n \u003cp\u003eOutcome\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eCC\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eMICE\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003cth align=\"left\"\u003e\u0026nbsp;\u003c/th\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eN\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eBETA (85% CI)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eP\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eN\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eBETA (85% CI)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eP\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eN\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eBETA (85% CI)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eP\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eN\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eBETA (85% CI)\u003c/p\u003e\n \u003c/th\u003e\n \u003cth align=\"left\"\u003e\n \u003cp\u003eP\u003c/p\u003e\n \u003c/th\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003emTOR/AMPK\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGlucose, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.54 (-1.35, 2.43)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.68\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.38 (-1.96, 2.72)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.82\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.51 (-1.23, 2.25)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.45 (-1.86, 2.77)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.78\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eInsulin, pmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.32 (-1.53, 0.89)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.24 (-1.30, 0.81)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.32 (-1.54, 0.90)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.24 (-1.30, 0.82)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHOMA-IR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.04 (-0.43, 0.34)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.87\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.07 (-0.40, 0.26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.05 (-0.44, 0.34)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.06 (-0.39, 0.27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.79\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAdiponectin, ug/ml\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-2.16 (-4.38, 0.06)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.16\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-2.12 (-4.46, 0.22)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-2.11 (-4.34, 0.11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-2.18 (-4.52, 0.17)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.18\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLeptin, \u0026micro;g/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.37 (-1.21, 0.48)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.03 (-0.83, 0.77)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.38 (-1.23, 0.47)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.44\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.04 (-0.84, 0.77)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.94\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCholesterol, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3.11 (-3.30, 9.51)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.24 (-5.39, 7.88)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.79\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3.19 (-3.22, 9.59)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.35 (-5.29, 7.98)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.77\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTriglycerides, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.58 (-7.46, 8.62)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-2.30 (-10.89, 6.29)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.48 (-7.57, 8.53)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-2.37 (-10.95, 6.21)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.69\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHDL, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.56 (-1.97, 5.09)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.93 (-3.34, 5.20)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.63 (-1.85, 5.11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.02 (-3.23, 5.27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.73\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLDL, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-5.07 (-12.07, 1.92)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.29\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-3.59 (-9.38, 2.20)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-5.04 (-12.14, 2.07)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-3.59 (-9.42, 2.23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.37\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e%Fat mass\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-2.44 (-3.85, -1.02)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.01*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-2.44 (-3.83, -1.04)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.01*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-2.44 (-3.87, -1.01)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.01*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-2.44 (-3.85, -1.03)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.01*\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBMI percentile\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-1.77 (-7.15, 3.61)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-1.77 (-7.08, 3.55)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-1.86 (-7.26, 3.54)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-1.86 (-7.20, 3.47)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.62\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eIGF/Mitochondrial Biogenesis\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGlucose, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.81 (-0.75, 2.38)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.45\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.86 (-1.27, 2.99)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.56\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.76 (-0.68, 2.21)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.75 (-1.37, 2.86)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.61\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eInsulin, pmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.22 (-1.04, 1.48)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.42 (-0.54, 1.37)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.22 (-1.05, 1.49)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.77\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.42 (-0.54, 1.38)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.53\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHOMA-IR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.04 (-0.39, 0.32)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.88\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.07 (-0.36, 0.21)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.03 (-0.38, 0.33)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.92\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.08 (-0.37, 0.21)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.68\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAdiponectin, ug/ml\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.64 (-0.51, 3.79)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.27\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.98 (-1.33, 3.29)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.54\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.63 (-0.52, 3.79)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.23\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.07 (-1.24, 3.38)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.51\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLeptin, \u0026micro;g/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.68 (-0.36, 1.72)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.36 (-0.74, 1.45)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.64\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.68 (-0.37, 1.73)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.37 (-0.71, 1.45)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.62\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCholesterol, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.70 (-3.35, 6.74)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.63\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.07 (-3.67, 5.82)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.33 (-3.76, 6.41)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.92 (-3.82, 5.66)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.78\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTriglycerides, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e11.62 (5.85, 17.39)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.01*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e8.34 (1.83, 14.84)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.07~\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e12.25 (6.51, 18.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.01*\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e8.48 (1.88, 15.08)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.06~\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHDL, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.72 (-2.06, 3.50)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.78 (-2.74, 4.29)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.75\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.40 (-2.36, 3.17)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.64 (-2.86, 4.14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.79\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLDL, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.20 (-3.71, 8.12)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.59\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.77 (-6.45, 4.92)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.26 (-3.75, 8.27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.61\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.77 (-6.49, 4.95)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.85\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e%Fat mass\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.53 (-0.76, 1.81)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.53 (-0.74, 1.80)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.55\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.53 (-0.77, 1.83)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.53 (-0.75, 1.81)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.55\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBMI percentile\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3.00 (-1.62, 7.63)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3.00 (-1.57, 7.57)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.34\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3.17 (-1.48, 7.81)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3.17 (-1.42, 7.76)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.32\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003ePlacental Insulin Coordination\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGlucose, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.15 (-4.07, 3.76)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.95\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-2.80 (-6.68, 1.07)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-1.56 (-5.22, 2.11)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-3.79 (-7.58, -0.01)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.15\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eInsulin, pmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-3.01 (-5.71, -0.30)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.11~\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-2.31 (-4.48, -0.14)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.12~\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-3.02 (-5.78, -0.27)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.10~\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-2.40 (-4.62, -0.18)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.12~\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHOMA-IR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.97 (-1.81, -0.13)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.10~\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.22 (-1.47, 1.04)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-1.10 (-1.95, -0.26)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.06~\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.29 (-1.53, 0.96)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAdiponectin, ug/ml\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.59 (-3.82, 7.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.47 (-2.84, 7.79)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.13 (-3.34, 7.59)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.70\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3.19 (-2.37, 8.75)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.41\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLeptin, \u0026micro;g/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.89 (-2.78, 1.01)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.65 (-2.52, 1.23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.62\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.75 (-2.70, 1.20)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.65\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.57 (-2.56, 1.42)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.68\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCholesterol, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e10.25 (-2.24, 22.74)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.24\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e8.07 (-3.34, 19.48)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e9.43 (-3.17, 22.03)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e7.23 (-4.25, 18.70)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.36\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTriglycerides, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.79 (-13.05, 18.63)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.80\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e9.26 (-14.06, 32.58)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3.97 (-11.97, 19.91)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e10.45 (-12.81, 33.72)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.52\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHDL, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.67 (-4.29, 9.64)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.58\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.15 (-8.89, 8.60)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.93 (-4.99, 8.85)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-1.18 (-9.95, 7.59)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.85\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLDL, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.36 (-15.97, 18.68)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3.65 (-12.35, 19.65)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.89 (-15.88, 19.67)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.85\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e3.80 (-12.39, 20.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e%Fat mass\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.19 (-0.69, 5.07)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.27\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.19 (-0.65, 5.04)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.27\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.27 (-0.68, 5.23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e2.27 (-0.65, 5.19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.26\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBMI percentile\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e9.85 (-0.66, 20.36)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.18\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e9.85 (-0.54, 20.24)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.17\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e11.35 (0.64, 22.06)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.08~\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e11.35 (0.77, 21.94)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.12~\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e\u003cstrong\u003eInflammation/Stress\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003ctd align=\"left\"\u003e\u0026nbsp;\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eGlucose, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.17 (-2.60, 2.93)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.24 (-4.45, 3.98)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.51 (-3.08, 2.06)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.71\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.59 (-4.84, 3.66)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.84\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eInsulin, pmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.50 (-0.63, 3.63)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.22 (-0.56, 3.01)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e52\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.56 (-0.62, 3.73)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.27\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.23 (-0.58, 3.04)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.33\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHOMA-IR\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.44 (-0.11, 0.99)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.30 (-0.32, 0.92)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.49\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e36\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.40 (-0.16, 0.96)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.30\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.28 (-0.35, 0.91)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.52\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eAdiponectin, ug/ml\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.21 (-3.86, 4.28)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.94\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.76 (-4.40, 5.92)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e51\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.51 (-3.59, 4.61)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.74\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e1.02 (-4.20, 6.23)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.78\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLeptin, \u0026micro;g/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.90 (-2.38, 0.57)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.63 (-2.23, 0.97)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.57\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e50\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.82 (-2.33, 0.68)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.31\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.59 (-2.24, 1.05)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.61\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eCholesterol, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-7.34 (-16.51, 1.84)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.25\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-7.34 (-15.90, 1.22)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-8.90 (-18.19, 0.39)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.22\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-7.87 (-16.54, 0.80)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.19\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eTriglycerides, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-7.45 (-18.97, 4.07)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.35\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-4.55 (-17.12, 8.03)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.60\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-6.18 (-17.98, 5.61)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.32\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-4.26 (-17.02, 8.51)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.63\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eHDL, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.40 (-4.73, 5.53)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.91\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.10 (-5.07, 4.87)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e48\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.63 (-5.79, 4.52)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.83\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.51 (-5.52, 4.50)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.88\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eLDL, mmol/l\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-12.73 (-22.46, -3.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.06~\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-11.24 (-23.55, 1.07)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.19\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e37\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-12.71 (-22.70, -2.72)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.06~\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-11.33 (-23.86, 1.19)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.19\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e%Fat mass\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.05 (-2.63, 2.53)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.05 (-2.60, 2.50)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.98\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.05 (-2.67, 2.57)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.90\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.05 (-2.63, 2.54)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.98\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003eBMI percentile\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.33 (-9.46, 8.80)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e-0.33 (-9.36, 8.69)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.96\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.11 (-9.09, 9.31)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.87\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e72\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"left\"\u003e\n \u003cp\u003e0.11 (-8.99, 9.20)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd align=\"char\"\u003e\n \u003cp\u003e0.99\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003ctfoot\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"13\"\u003eNotes: Missing data were imputed using multiple imputation by chained equations (MICE), generating five imputed datasets. The imputation model included maternal and child predictors, with predictive mean matching for continuous variables and logistic regression for categorical variables. Estimates represent pooled regression coefficients across imputations, with standard errors calculated using Rubin\u0026rsquo;s rules. Confidence intervals and p-values were derived using a normal approximation (z-test), rather than the default t-distribution approach.\u003cbr\u003e\n \u003cp\u003eAbbreviations: AMPK, AMP-activated protein kinase; mTOR, Mechanistic Target of Rapamycin, IGF, Insulin Like Growth Factor; HOMA-IR, Homeostatic Model Assessment of Insulin Resistance; HDL, high-density lipoprotein; LDL, low-density lipoprotein; TGs, Triglycerides; BMI, body mass index;\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e*P-value \u0026lt;0.05 (95% confidence); ~ P-value\u0026lt;0.15 (85% confidence)\u003c/p\u003e\n \u003cp\u003eModel 1: Age and sex\u003c/p\u003e\n \u003cp\u003eModel 2 (Maternal): pre-pregnancy body mass index (kg/m2)\u003c/p\u003e\n \u003cp\u003eModels of neonatal fat mass % are reported in Supplemental Table 2.\u003c/p\u003e\u003cbr\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tfoot\u003e\n \u003c/table\u003e\n \u003c/div\u003e\n \u003cp\u003e\u003cem\u003eSensitivity analysis\u003c/em\u003e\u003c/p\u003e\n \u003cp\u003eIn sensitivity analyses the dimensions retained were restricted to the top three consensus clusters based on size (\u0026ldquo;\u003cem\u003ePlacental Insulin Coordination\u003c/em\u003e\u0026rdquo;, \u0026ldquo;\u003cem\u003emTOR/AMPK\u003c/em\u003e\u0026rdquo;, \u0026ldquo;\u003cem\u003eInflammation/Stress\u003c/em\u003e\u0026rdquo;), the top three PCs based on variance explained (1 through 3), and the top three WGCNA modules based on mean module connectivity (\u0026ldquo;brown\u0026rdquo;, \u0026ldquo;turquoise\u0026rdquo;, \u0026ldquo;blue\u0026rdquo;). In these analyses, PCs performed slightly better for neonatal fat mass percentage; however, for childhood fat mass percentage, consensus clustering remained the best model (\u003cstrong\u003eSupplemental Table 1\u003c/strong\u003e). In general, models that mutually adjusted for all four consensus clusters had similar results in association with neonatal and childhood fat mass to those that assessed each cluster individually (\u003cstrong\u003eSupplemental Table 2\u003c/strong\u003e), suggesting independence of the pathways captured by each cluster. We utilized separate models for each cluster because our interest was to identify associations with clusters of proteins, without holding constant the activity of the other proteins measured and assigned to different clusters.\u0026nbsp;\u003c/p\u003e\n\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eIn this pregnancy cohort with longitudinal follow-up from birth through early childhood (median offspring age 4.79 ± 0.6 years), we compared three data reduction methods—consensus clustering, WGCNA, and PCA—for capturing placental signaling across multiple pathways potentially involved in in utero programming of offspring cardiometabolic health. Based on model fit metrics, consensus clustering emerged as the optimal for predicting neonatal and early childhood fat mass percentage. The clusters, which captured signaling in pathways of mTOR, AMPK, IGF/insulin signaling, mitochondrial biogenesis, and inflammation and stress responses were also associated with offspring cardiometabolic outcomes at four years of age.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIn our comparison of the data reduction approaches, the general modest correlation structure observed in the placental protein data (median pairwise Spearman correlation 0.13; IQR: 0.06–0.22) may have influenced the relative performance of the data reduction methods in our analysis. Consensus clustering outperformed PCA and WGCNA in the models using all reduced dimensions, likely due to its ability to detect localized co-variation (e.g., smaller sets of variation) without relying on strong global correlation. Each approach has distinct assumptions about the underlying structure of [biological] data, particularly with respect to correlation. PCA reduces dimensionality by identifying orthogonal axes of shared variance and retains the global correlation structure, allowing each protein to contribute to multiple components.(50) WGCNA also relies on global correlation, using pairwise relationships to identify modules of highly correlated proteins.(29) In contrast, consensus clustering does not explicitly rely on correlation; instead, it uses repeated resampling to identify stable groupings of proteins that co-vary across subsets of samples,(28) which may capture more nuanced groupings of proteins that are co-regulated. While consensus clustering of placental protein data resulted in the best model fit and the highest explanatory power, the best model fit will vary depending on the number and types of proteins measured, as well as the specific outcome. We present these three complementary approaches and their comparisons as a guide for others seeking an alternative to single protein models. Whether consensus clustering is an optimal approach that could be adopted more generally can only be determined with replication in other study populations.\u003c/p\u003e\n\u003cp\u003eWe found that the “\u003cem\u003emTOR/AMPK\u003c/em\u003e” cluster, characterized by increased activation of RPS6, 4E-BP, and AMPK, was inversely associated with neonatal and early childhood fat mass percentage. This cluster also showed a general inverse pattern with the other metabolic markers during early childhood, although there was imprecision in these models. These overall findings suggest a potential link between this placental signaling pattern and lower offspring adiposity, with possible implications for longer-term metabolic health. These findings align with prior research reporting inverse associations of AMPK activation and birthweight,(51, 52) as well as broader evidence linking AMPK activity with improved metabolic outcomes.(53) Importantly, most prior studies on placental mTOR and AMPK signaling were conducted in high-risk pregnancies (e.g., maternal obesity, hyperglycemia, or hypoxia), where placental function is often pathologically disrupted.(25, 52, 54-56) In contrast, our cohort primarily comprised of low-risk pregnancies, allowed us to observe more physiological—and possibly adaptive—signaling patterns, rather than pathological ones.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eA particularly notable aspect of the “\u003cem\u003emTOR/AMPK\u003c/em\u003e” cluster was the co-activation of AMPK with mTORC1 downstream targets RPS6 and 4E-BP1. This may appear counterintuitive, as AMPK is classically thought to inhibit mTOR signaling, particularly in energy-stressed states.(57) However, placental signaling does not always follow traditional nutrient-sensing paradigms observed in liver, adipose, or skeletal muscle. Indeed, prior animal studies have shown that placental mTOR can be inhibited even in the absence of AMPK activation, suggesting distinct regulatory mechanisms.(22, 58) Co-activation of AMPK and mTOR effectors may reflect a placenta-specific strategy to balance energy sensing with protein synthesis, in the context of supporting both placental and fetal growth.(59) For example, we have previously speculated that 4E-BP1 activation may preferentially support placental growth and biosynthesis, whereas RPS6 signaling may be more directly involved in nutrient transport, oxidative metabolism, and fetal growth.(31) In the context of our study, the clustering of AMPK, 4E-BP1, and RPS6—along with their inverse association with neonatal fat mass— may indicate a coordinated signaling state to balance fetal growth priorities and placental function to prevent excess nutrient transfer to the fetus. However, this remains a hypothesis that requires further mechanistic investigation.\u0026nbsp; Given the complex regulation of placental signaling by several inputs (e.g., glucose, amino acids, blood flow, oxygenation), these data underscore the need for future studies to disentangle how integrated placental signaling networks shape fetal nutrient exposure and influence early metabolic trajectories.(60)\u003c/p\u003e\n\u003cp\u003eWe found that “\u003cem\u003eIGF/Mitochondria Biogenesis\u003c/em\u003e” cluster—marked by higher IGF-1r and PGC1α—was positively associated with childhood triglyceride levels. While adjustment for child’s lifestyle modestly attenuated this association and the confidence intervals included the null, the pattern remained directionally consistent. The clustering of IGF-1r and PGC1α is biologically plausible, as IGF-1r is a known upstream regulator of PGC1α expression and mitochondrial biogenesis.(61) Beyond the role of PGC1α in mitochondrial remodeling, it also influences lipid metabolism including long-chain fatty acid oxidation and hepatic triglyceride storage and secretion.(62, 63) Taken together, these findings suggest that enhanced placental IGF-1 signaling and mitochondrial activity may prime lipid metabolic pathways in utero, potentially increasing susceptibility to elevated triglycerides in childhood. The attenuation of associations after accounting for child lifestyle exposures may reflect an interaction between early programing and postnatal obesogenic environments, potentially amplifying the effects of early programming through a secondary “second-hit” mechanism.(64)\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe observed that the “\u003cem\u003ePlacental Insulin Coordination”\u0026nbsp;\u003c/em\u003ecluster—which included phosphorylated proteins across insulin signaling (IRβ, ERK1/2), mTOR nutrient transport (total RPS6), and stress-response pathways (JNK)—was modestly associated with lower HOMA-IR in early childhood. These associations were not attenuated by adjustment for maternal BMI or child lifestyle. While the clustering of activated proteins from these different pathways may seem inconsistent with canonical nutrient-sensing models, the clustering pattern may represent coordinated crosstalk among insulin, nutrient-sensing, and stress-response pathways that maintain placental homeostasis in low-risk pregnancies. Mechanistically, insulin receptor signaling activates both mitogenic (ERFK1/2) and metabolic (AKT) arms, with feedback via mTOR and JNK that modulate the intensity of downstream signals.(65, 66) As such, the clustering of activated ERK1/2 and JNK with total RPS6 may indicate a more responsive and balanced placental signaling state—potentially enhancing fetal metabolic programing and contributing to improved insulin sensitivity in offspring.\u003c/p\u003e\n\u003cp\u003eThe “\u003cem\u003eInflammation/Stress\u003c/em\u003e” cluster, comprised of proteins from MAPK, inflammatory signaling, and stress response signaling (ERK1/2, p38MAPK, STAT3, JNK1, and IL-1β), was inversely associated with childhood LDL cholesterol levels. This association was slightly attenuated after adjustment for maternal BMI and, to a greater extent, after accounting for child lifestyle factors. Given this cluster primarily reflected total rather than phosphorylated protein levels, it may capture baseline signaling capacity or shared maternal-child metabolic milieu rather than an independent contributor to offspring cardiometabolic health.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStrengths and Limitations\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA major strength of this study is the use of absolute quantification for a panel of placental proteins involved in key signaling and nutrient transport pathways; and the measurement of protein phosphorylation at specific serine and tyrosine residues that indicate activation. This approach allowed for direct comparison of expression levels and pathway activation across proteins and participants, enhancing biological interpretability. Another strength is the focus on a generally healthy pregnancy cohort, with a low prevalence of GDM (n=5) and maternal obesity (n=13). This low-risk context enabled identification of protein clustering patterns that likely reflect physiological coordination rather than pathologic dysregulation.\u003c/p\u003e\n\u003cp\u003eThis focus is especially important given that many canonical signaling relationships observed in other metabolic tissues—such as liver, adipose, or skeletal muscle—do not always translate directly to the placenta. By analyzing protein clusters in a healthy cohort, we were able to examine signaling patterns that may reflect normal physiological coordination of placental function, an area often underexplored in favor of pathophysiologic models. This perspective is particularly valuable in light of the limited foundational knowledge about normal signaling regulation in the human placenta.(67)\u003c/p\u003e\n\u003cp\u003eHowever, these strengths must be considered alongside limitations. The clustering patterns and their associations with offspring outcomes observed in this study may not generalize to populations with greater metabolic or obstetric risk. Future work is needed to examine whether these placental signaling profiles are preserved, amplified, or disrupted in pregnancies complicated by conditions such as obesity, diabetes, or hypertension. We acknowledge that while consensus clustering provided the most robust model fit in this context, all data reduction techniques involve trade-offs and may oversimplify complex biological relationship or obscure single protein-outcome associations. Thus, the use of data reduction versus traditional single protein models should be informed by the study objectives. In addition, the relatively small sample size may have reduced the precision of effect estimates. To mitigate this, we used multiple imputation to complete missing data among those with follow-up, which maximized the sample size for all follow-up outcomes. Lastly, the study involved multiple tests of association between placental protein clusters and a range of childhood cardiometabolic outcomes. While this increases potential for false positives, we prioritized interpretation of results that were consistent across complete case and imputed models and robust to multivariable adjustment. For transparency, we present all models tested and have highlighted that findings should be considered hypothesis-generating, and none should be regarded as conclusive evidence.(68)\u0026nbsp;\u003c/p\u003e"},{"header":"Conclusions","content":"\u003cp\u003eWe identified distinct placental signaling patterns using consensus clustering, WGCNA, and PCA. Among these, latent signaling patterns derived from consensus clustering best predicted neonatal fat mass percentage. When linking the clusters of interest to offspring metabolic outcomes at median age 4.8 years, clusters that captured signaling across several pathways of\u0026nbsp;insulin/growth factor signaling, stress/inflammation, and mitochondrial biogenesis were associated with fat mass, HOMA-IR, and triglycerides. These findings highlight the value of studying coordinated placental signaling in shaping early metabolic health trajectories. While these associations should be interpreted in light of the study’s modest sample size and observational design, the results lay a foundation for future mechanistic research that could inform etiological insight and early-life predictors of long-term cardiometabolic risk.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eRPS6, Ribosomal protein S6\u003c/p\u003e\n\u003cp\u003eS6K1, p70 ribosomal protein S6 kinase 1\u003c/p\u003e\n\u003cp\u003e4E-BP1, Eukaryotic translation initiation factor 4E-binding protein 1\u003c/p\u003e\n\u003cp\u003eAkt, Protein kinase B\u003c/p\u003e\n\u003cp\u003ePKCα, Protein kinase C-alpha\u003c/p\u003e\n\u003cp\u003eIRβ, Insulin receptor beta\u003c/p\u003e\n\u003cp\u003eIGF-1r, Insulin-like growth factor 1 receptor\u003c/p\u003e\n\u003cp\u003eGSK3β, Glycogen synthase kinase-3 beta\u003c/p\u003e\n\u003cp\u003eERK, Extracellular signal-regulated kinase\u003c/p\u003e\n\u003cp\u003ePro-caspase 1, Pro-caspase 1\u003c/p\u003e\n\u003cp\u003eSTAT3, Signal transducer and activator of transcription 3\u003c/p\u003e\n\u003cp\u003eJNK1/2, c-Jun N-terminal kinase 1 and 2\u003c/p\u003e\n\u003cp\u003eIL-1β, Interleukin-1 beta\u003c/p\u003e\n\u003cp\u003ep38 MAPK, p38 mitogen-activated protein kinase\u003c/p\u003e\n\u003cp\u003eAMPK, AMP-activated protein kinase\u003c/p\u003e\n\u003cp\u003eeIF2α, Eukaryotic initiation factor 2 alpha\u003c/p\u003e\n\u003cp\u003eOGT, O-linked N-acetylglucosamine transferase\u003c/p\u003e\n\u003cp\u003ePI3K, Phosphoinositide 3-kinase\u003c/p\u003e\n\u003cp\u003e11β-HSD1, 11β-hydroxysteroid dehydrogenase type 1\u003c/p\u003e\n\u003cp\u003ePGC-1α, Peroxisome proliferator-activated receptor gamma coactivator 1-alpha\u003c/p\u003e\n\u003cp\u003emTOR, Mechanistic Target of Rapamycin\u0026nbsp;\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate: \u003c/strong\u003eThe Healthy Start Study (ClinicalTrials.gov; NCT02273297) was approved by the Colorado Multiple Institutional Review Board at the University of Colorado Hospital. All participants gave written informed consent at enrollment. \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication: \u003c/strong\u003eNot applicable \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials: \u003c/strong\u003eCode can be found at https://github.com/Ecfrancis13/unsupervised-data-reduction. The datasets generated and/or analysed during the current study are not publicly available due to informed consent, but are available on reasonable request at https://healthystartstudy.org/for-investigators \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests: \u003c/strong\u003eThe authors declare that they have no competing interests\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding: \u003c/strong\u003eThis research was funded by the Eunice Kennedy Shriver National Institute of Child Health and Human Development grant number R00 HD108272 and the National Institute of Diabetes, Digestive, and Kidney Diseases R01 DK068001\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors' contributions: \u003c/strong\u003eECF conceptualized the study, analyzed the data, and drafted the manuscript. TJ supervised the placental collection, protein assays, and provided critical review of the manuscript and revisions. DD conceptualized oversees the broader Healthy Start Study and provided critical review of the manuscript. WP, KEB, and LEG edited the manuscript, supported the interpretation of results, and provided critical review of the final version. \u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements: \u003c/strong\u003eWe thank the Healthy Start participants and staff for the valuable contributions to the study. \u003cstrong\u003e\u003cbr\u003e \u003c/strong\u003e\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eBurton GJ, Fowden AL, Thornburg KL. Placental Origins of Chronic Disease. Physiological reviews. 2016;96(4):1509-65.\u003c/li\u003e\n\u003cli\u003eBrett K, Ferraro Z, Holcik M, Adamo K. Prenatal physical activity and diet composition affect the expression of nutrient transporters and mTOR signaling molecules in the human placenta. Placenta. 2015;36(2):204-12.\u003c/li\u003e\n\u003cli\u003eThakali KM, Zhong Y, Cleves M, Andres A, Shankar K. Associations between maternal body mass index and diet composition with placental DNA methylation at term. Placenta. 2020;93:74-82.\u003c/li\u003e\n\u003cli\u003eKyriakis JM, Avruch J. Mammalian MAPK Signal Transduction Pathways Activated by Stress and Inflammation: A 10-Year Update. Physiological reviews. 2012;92(2):689-737.\u003c/li\u003e\n\u003cli\u003eWesolowski SR, Hay WW, Jr. Role of placental insufficiency and intrauterine growth restriction on the activation of fetal hepatic glucose production. Mol Cell Endocrinol. 2016;435:61-8.\u003c/li\u003e\n\u003cli\u003eSinghal A. Does early growth affect long-term risk factors for cardiovascular disease? Nestle Nutr Workshop Ser Pediatr Program. 2010;65:55-64; discussion -9.\u003c/li\u003e\n\u003cli\u003eMonteiro PO, Victora CG. Rapid growth in infancy and childhood and obesity in later life--a systematic review. Obes Rev. 2005;6(2):143-54.\u003c/li\u003e\n\u003cli\u003eBaird J, Fisher D, Lucas P, Kleijnen J, Roberts H, Law C. Being big or growing fast: systematic review of size and growth in infancy and later obesity. BMJ. 2005;331(7522):929.\u003c/li\u003e\n\u003cli\u003eFreedman DS, Dietz WH, Srinivasan SR, Berenson GS. The relation of overweight to cardiovascular risk factors among children and adolescents: the Bogalusa Heart Study. Pediatrics. 1999;103(6 Pt 1):1175-82.\u003c/li\u003e\n\u003cli\u003eFreedman DS, Khan LK, Serdula MK, Dietz WH, Srinivasan SR, Berenson GS. The relation of childhood BMI to adult adiposity: the Bogalusa Heart Study. Pediatrics. 2005;115(1):22-7.\u003c/li\u003e\n\u003cli\u003eJuhola J, Magnussen CG, Viikari JSA, K\u0026auml;h\u0026ouml;nen M, Hutri-K\u0026auml;h\u0026ouml;nen N, Jula A, et al. Tracking of Serum Lipid Levels, Blood Pressure, and Body Mass Index from Childhood to Adulthood: The Cardiovascular Risk in Young Finns Study. The Journal of pediatrics. 2011;159(4):584-90.\u003c/li\u003e\n\u003cli\u003eZinner SH, Rosner B, Oh W, Kass EH. Significance of blood pressure in infancy. Familial aggregation and predictive effect on later blood pressure. Hypertension. 1985;7(3 Pt 1):411-6.\u003c/li\u003e\n\u003cli\u003eWard ZJ, Long MW, Resch SC, Giles CM, Cradock AL, Gortmaker SL. Simulation of Growth Trajectories of Childhood Obesity into Adulthood. New England Journal of Medicine. 2017;377(22):2145-53.\u003c/li\u003e\n\u003cli\u003eIb\u0026aacute;\u0026ntilde;ez L, Ong K, Dunger DB, de Zegher F. Early development of adiposity and insulin resistance after catch-up weight gain in small-for-gestational-age children. The Journal of clinical endocrinology and metabolism. 2006;91(6):2153-8.\u003c/li\u003e\n\u003cli\u003eJuhola J, Magnussen CG, Viikari JS, Kahonen M, Hutri-Kahonen N, Jula A, et al. Tracking of serum lipid levels, blood pressure, and body mass index from childhood to adulthood: the Cardiovascular Risk in Young Finns Study. J Pediatr. 2011;159(4):584-90.\u003c/li\u003e\n\u003cli\u003eHampl SE, Hassink SG, Skinner AC, Armstrong SC, Barlow SE, Bolling CF, et al. Clinical Practice Guideline for the Evaluation and Treatment of Children and Adolescents With Obesity. Pediatrics. 2023;151(2).\u003c/li\u003e\n\u003cli\u003eChen L, Guilmette J, Luo ZC, Cloutier A, Wang WJ, Yang MN, et al. Placental 11beta-HSD2 and Cardiometabolic Health Indicators in Infancy. Diabetes Care. 2019;42(5):964-71.\u003c/li\u003e\n\u003cli\u003eKeleher MR, Erickson K, Kechris K, Yang IV, Dabelea D, Friedman JE, et al. Associations between the activity of placental nutrient-sensing pathways and neonatal and postnatal metabolic health: the ECHO Healthy Start cohort. Int J Obes (Lond). 2020.\u003c/li\u003e\n\u003cli\u003eKeleher MR, Erickson K, Smith HA, Kechris KJ, Yang IV, Dabelea D, et al. Placental Insulin/IGF-1 Signaling, PGC-1alpha, and Inflammatory Pathways Are Associated With Metabolic Outcomes at 4-6 Years of Age: The ECHO Healthy Start Cohort. Diabetes. 2021;70(3):745-51.\u003c/li\u003e\n\u003cli\u003eGardebjer EM, Cuffe JS, Pantaleon M, Wlodek ME, Moritz KM. Periconceptional alcohol consumption causes fetal growth restriction and increases glycogen accumulation in the late gestation rat placenta. Placenta. 2014;35(1):50-7.\u003c/li\u003e\n\u003cli\u003eSferruzzi-Perri AN, Sandovici I, Constancia M, Fowden AL. Placental phenotype and the insulin-like growth factors: resource allocation to fetal growth. J Physiol. 2017;595(15):5057-93.\u003c/li\u003e\n\u003cli\u003eRosario FJ, Jansson N, Kanai Y, Prasad PD, Powell TL, Jansson T. Maternal Protein Restriction in the Rat Inhibits Placental Insulin, mTOR, and STAT3 Signaling and Down-Regulates Placental Amino Acid Transporters. Endocrinology. 2011;152(3):1119-29.\u003c/li\u003e\n\u003cli\u003eJansson T, Powell TL. Human Placental Transport in Altered Fetal Growth: Does the Placenta Function as a Nutrient Sensor? \u0026ndash; A Review. Placenta. 2006;27:91-7.\u003c/li\u003e\n\u003cli\u003ePantham P, Rosario FJ, Weintraub ST, Nathanielsz PW, Powell TL, Li C, et al. Down-Regulation of Placental Transport of Amino Acids Precedes the Development of Intrauterine Growth Restriction in Maternal Nutrient Restricted Baboons. Biology of reproduction. 2016;95(5):98.\u003c/li\u003e\n\u003cli\u003eDumolt JH, Powell TL, Jansson T. Placental Function and the Development of Fetal Overgrowth and Fetal Growth Restriction. Obstet Gynecol Clin North Am. 2021;48(2):247-66.\u003c/li\u003e\n\u003cli\u003eCalejman CM, Doxsey WG, Fazakerley DJ, Guertin DA. Integrating adipocyte insulin signaling and metabolism in the multi-omics era. Trends in Biochemical Sciences. 2022;47(6):531-46.\u003c/li\u003e\n\u003cli\u003ePaquette AG, Hood L, Price ND, Sadovsky Y. Deep phenotyping during pregnancy for predictive and preventive medicine. Science Translational Medicine. 2020;12(527):eaay1059.\u003c/li\u003e\n\u003cli\u003eWilkerson MD, Hayes DN. ConsensusClusterPlus: a class discovery tool with confidence assessments and item tracking. Bioinformatics. 2010;26(12):1572-3.\u003c/li\u003e\n\u003cli\u003eLangfelder P, Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics. 2008;9(1):559.\u003c/li\u003e\n\u003cli\u003eHarrod CS, Fingerlin TE, Chasan-Taber L, Reynolds RM, Glueck DH, Dabelea D. Exposure to prenatal smoking and early-life body composition: the healthy start study. Obesity (Silver Spring). 2015;23(1):234-41.\u003c/li\u003e\n\u003cli\u003eKeleher MR, Erickson K, Kechris K, Yang IV, Dabelea D, Friedman JE, et al. Associations between the activity of placental nutrient-sensing pathways and neonatal and postnatal metabolic health: the ECHO Healthy Start cohort. International Journal of Obesity. 2020.\u003c/li\u003e\n\u003cli\u003eStarling AP, Brinton JT, Glueck DH, Shapiro AL, Harrod CS, Lynch AM, et al. Associations of maternal BMI and gestational weight gain with neonatal adiposity in the Healthy Start study. The American Journal of Clinical Nutrition. 2015;101(2):302-9.\u003c/li\u003e\n\u003cli\u003eFrancis EC, Dabelea D, Shankar K, Perng W. Maternal diet quality during pregnancy is associated with biomarkers of metabolic risk among male offspring. Diabetologia. 2021.\u003c/li\u003e\n\u003cli\u003eInstitute of Medicine. Weight gain during pregnancy: reexamining the guidelines. Washington, DC; 2009.\u003c/li\u003e\n\u003cli\u003eSubar AF, Kirkpatrick SI, Mittl B, Zimmerman TP, Thompson FE, Bingley C, et al. The Automated Self-Administered 24-hour dietary recall (ASA24): a resource for researchers, clinicians, and educators from the National Cancer Institute. J Acad Nutr Diet. 2012;112(8):1134-7.\u003c/li\u003e\n\u003cli\u003eShapiro AL, Kaar JL, Crume TL, Starling AP, Siega-Riz AM, Ringham BM, et al. Maternal diet quality in pregnancy and neonatal adiposity: the Healthy Start Study. International journal of obesity (2005). 2016;40(7):1056-62.\u003c/li\u003e\n\u003cli\u003eWest KA, Schmid R, Gauglitz JM, Wang M, Dorrestein PC. foodMASST a mass spectrometry search tool for foods and beverages. npj Science of Food. 2022;6(1):22.\u003c/li\u003e\n\u003cli\u003eFrancis EC, Dabelea D, Boyle KE, Jansson T, Perng W. Maternal Diet Quality Is Associated with Placental Proteins in the Placental Insulin/Growth Factor, Environmental Stress, Inflammation, and mTOR Signaling Pathways: The Healthy Start ECHO Cohort. J Nutr. 2022;152(3):816-25.\u003c/li\u003e\n\u003cli\u003eGuenther PM, Kirkpatrick SI, Reedy J, Krebs-Smith SM, Buckman DW, Dodd KW, et al. The Healthy Eating Index-2010 is a valid and reliable measure of diet quality according to the 2010 Dietary Guidelines for Americans. The Journal of nutrition. 2014;144(3):399-407.\u003c/li\u003e\n\u003cli\u003eGuenther PM, Casavale KO, Reedy J, Kirkpatrick SI, Hiza HA, Kuczynski KJ, et al. Update of the Healthy Eating Index: HEI-2010. J Acad Nutr Diet. 2013;113(4):569-80.\u003c/li\u003e\n\u003cli\u003eFields DA, Allison DB. Air-Displacement Plethysmography Pediatric Option in 2\u0026ndash;6 Years Old Using the Four-Compartment Model as a Criterion Method. Obesity. 2012;20(8):1732-7.\u003c/li\u003e\n\u003cli\u003eWorld Health Organizaton. WHO child growth standards: height-for-age, weight-for-age, weight-for-length, weight-for-height and body mass index for-age: methods and development. Geneva, Switzerland: World Health Organizaton; 2006.\u003c/li\u003e\n\u003cli\u003ede Onis M, Garza C, Victora C, Bhan M, Norum K. The WHO Multicentre Growth Reference Study (MGRS): rationale, planning, and implementaton. Food Nutr Bull. 2004;25(Suppl1):S1-S89.\u003c/li\u003e\n\u003cli\u003eHediger ML, Overpeck MD, Ruan WJ, Troendle JF. Early infant feeding and growth status of US-born infants and children aged 4-71 mo: analyses from the third National Health and Nutrition Examination Survey, 1988-1994. The American journal of clinical nutrition. 2000;72(1):159-67.\u003c/li\u003e\n\u003cli\u003eRowlands AV. Accelerometer assessment of physical activity in children: an update. Pediatr Exerc Sci. 2007;19(3):252-66.\u003c/li\u003e\n\u003cli\u003evan Cauwenberghe E, Labarque V, Trost SG, de Bourdeaudhuij I, Cardon G. Calibration and comparison of accelerometer cut points in preschool children. International journal of pediatric obesity : IJPO : an official journal of the International Association for the Study of Obesity. 2011;6(2-2):e582-e9.\u003c/li\u003e\n\u003cli\u003eSteinberger J, Daniels SR, Hagberg N, Isasi CR, Kelly AS, Lloyd-Jones D, et al. Cardiovascular Health Promotion in Children: Challenges and Opportunities for 2020 and Beyond: A Scientific Statement From the American Heart Association. Circulation. 2016;134(12):e236-55.\u003c/li\u003e\n\u003cli\u003eOken E, Kleinman KP, Rich-Edwards J, Gillman MW. A nearly continuous measure of birth weight for gestational age using a United States national reference. BMC Pediatrics. 2003;3(1):6.\u003c/li\u003e\n\u003cli\u003eWHO Multicentre Growth Reference Study Group. WHO Child Growth Standards: Growth velocity based on weight, length and head circumference: Methods and development. Geneva: World Health Organization; 2009.\u003c/li\u003e\n\u003cli\u003eZwick RW, Velicer FW. Comparison of five rules for determining the number of components to retain. 1986;99:432-42.\u003c/li\u003e\n\u003cli\u003eLazo-de-la-Vega-Monroy M-L, Mata-Tapia K-A, Garcia-Santillan J-A, Corona-Figueroa M-A, Gonzalez-Dominguez M-I, Gomez-Zapata H-M, et al. Association of placental nutrient sensing pathways with birth weight. Reproduction. 2020;160(3):455-68.\u003c/li\u003e\n\u003cli\u003eJansson N, Rosario FJ, Gaccioli F, Lager S, Jones HN, Roos S, et al. Activation of Placental mTOR Signaling and Amino Acid Transporters in Obese Women Giving Birth to Large Babies. The Journal of Clinical Endocrinology \u0026amp; Metabolism. 2013;98(1):105-13.\u003c/li\u003e\n\u003cli\u003eZhang BB, Zhou G, Li C. AMPK: An Emerging Drug Target for Diabetes and the Metabolic Syndrome. Cell Metabolism. 2009;9(5):407-16.\u003c/li\u003e\n\u003cli\u003eDimasuay KG, Boeuf P, Powell TL, Jansson T. Placental Responses to Changes in the Maternal Environment Determine Fetal Growth. Frontiers in physiology. 2016;7:12-.\u003c/li\u003e\n\u003cli\u003eMa Y, Zhu MJ, Uthlaut AB, Nijland MJ, Nathanielsz PW, Hess BW, et al. Upregulation of growth signaling and nutrient transporters in cotyledons of early to mid-gestational nutrient restricted ewes. Placenta. 2011;32(3):255-63.\u003c/li\u003e\n\u003cli\u003eGaccioli F, White V, Capobianco E, Powell TL, Jawerbaum A, Jansson T. Maternal Overweight Induced by a Diet with High Content of Saturated Fat Activates Placental mTOR and eIF2alpha Signaling and Increases Fetal Growth in Rats. Biology of reproduction. 2013;89(4).\u003c/li\u003e\n\u003cli\u003eKola B, Grossman AB, Korbonits M, Korbonits M. The Role of AMP-Activated Protein Kinase in Obesity. Frontiers of hormone research. 2008;36:198-211.\u003c/li\u003e\n\u003cli\u003eKavitha JV, Rosario FJ, Nijland MJ, McDonald TJ, Wu G, Kanai Y, et al. Down-regulation of placental mTOR, insulin/IGF-I signaling, and nutrient transporters in response to maternal nutrient restriction in the baboon. FASEB journal : official publication of the Federation of American Societies for Experimental Biology. 2014;28(3):1294-305.\u003c/li\u003e\n\u003cli\u003eTsai K, Tullis B, Jensen T, Graff T, Reynolds P, Arroyo J. Differential expression of mTOR related molecules in the placenta from gestational diabetes mellitus (GDM), intrauterine growth restriction (IUGR) and preeclampsia patients. Reproductive Biology. 2021;21(2):100503.\u003c/li\u003e\n\u003cli\u003eKramer AC, Jansson T, Bale TL, Powell TL. Maternal-fetal cross-talk via the placenta: influence on offspring development and metabolism. Development. 2023;150(20).\u003c/li\u003e\n\u003cli\u003eLyons A, Coleman M, Riis S, Favre C, O\u0026apos;Flanagan CH, Zhdanov AV, et al. Insulin-like growth factor 1 signaling is essential for mitochondrial biogenesis and mitophagy in cancer cells. The Journal of biological chemistry. 2017;292(41):16983-98.\u003c/li\u003e\n\u003cli\u003eZhang Y, Castellani LW, Sinal CJ, Gonzalez FJ, Edwards PA. Peroxisome proliferator-activated receptor-gamma coactivator 1alpha (PGC-1alpha) regulates triglyceride metabolism by activation of the nuclear receptor FXR. Genes Dev. 2004;18(2):157-69.\u003c/li\u003e\n\u003cli\u003eLiang H, Ward WF. PGC-1alpha: a key regulator of energy metabolism. Adv Physiol Educ. 2006;30(4):145-51.\u003c/li\u003e\n\u003cli\u003eCheong JN, Wlodek ME, Moritz KM, Cuffe JS. Programming of maternal and offspring disease: impact of growth restriction, fetal sex and transmission across generations. J Physiol. 2016;594(17):4727-40.\u003c/li\u003e\n\u003cli\u003ePetersen MC, Shulman GI. Mechanisms of insulin action and insulin resistance. Physiological reviews. 2018.\u003c/li\u003e\n\u003cli\u003eLe TKC, Dao XD, Nguyen DV, Luu DH, Bui TMH, Le TH, et al. Insulin signaling and its application. Frontiers in Endocrinology. 2023;Volume 14 - 2023.\u003c/li\u003e\n\u003cli\u003eHerrera CL, Kim MJ, Do QN, Owen DM, Fei B, Twickler DM, et al. The human placenta project: Funded studies, imaging technologies, and future directions. Placenta. 2023;142:27-35.\u003c/li\u003e\n\u003cli\u003eRothman KJ. No adjustments are needed for multiple comparisons. Epidemiology (Cambridge, Mass). 1990;1(1):43-6.\u003c/li\u003e\n\u003cli\u003evan den Berg RA, Hoefsloot HCJ, Westerhuis JA, Smilde AK, van der Werf MJ. Centering, scaling, and transformations: improving the biological information content of metabolomics data. BMC Genomics. 2006;7(1):142.\u003c/li\u003e\n\u003cli\u003eRevelle W. psych: Procedures for Psychological, Psychometric, and Personality Research. Evanston, IL: Northwestern University; 2025.\u003c/li\u003e\n\u003cli\u003eHaggarty P. Placental regulation of fatty acid delivery and its effect on fetal growth--a review. Placenta. 2002;23 Suppl A:S28-38.\u003c/li\u003e\n\u003cli\u003eWagenmakers EJ, Farrell S. AIC model selection using Akaike weights. Psychon Bull Rev. 2004;11(1):192-6.\u003c/li\u003e\n\u003cli\u003eRubin DB. Multiple imputation for nonresponse in surveys. New York ;: Wiley; 1987. xxix, 258 p. p.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bmed","sideBox":"Learn more about [BMC Medicine](http://bmcmedicine.biomedcentral.com/)","snPcode":"12916","submissionUrl":"https://submission.nature.com/new-submission/12916/3","title":"BMC Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"placenta, maternal-fetal exchange, insulin signaling, nutrient sensing, fetal programming, cardiometabolic health, data reduction, unsupervised analysis, pregnancy","lastPublishedDoi":"10.21203/rs.3.rs-7926327/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7926327/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003ePlacental signaling pathways regulate nutrient transport and fetal growth, with potential long-term consequences for offspring metabolic health. Most prior human studies have focused on individual placental markers, limiting insight into the role of coordinated activity across multiple pathways in relation to offspring outcomes.\u003c/p\u003e\u003ch2\u003eObjective\u003c/h2\u003e\u003cp\u003eTo compare three unsupervised data reduction techniques for characterizing placental signaling patterns across multiple pathways and assess their associations with neonatal and early childhood adiposity and metabolic biomarkers.\u003c/p\u003e\u003ch2\u003eDesign:\u003c/h2\u003e\u003cp\u003eAmong 108 mother-child pairs from the Healthy Start cohort, we quantified 33 placental signaling proteins and their phosphorylated-to-total protein ratios involved in nutrient sensing, insulin/growth factor signaling, stress/inflammation, and mitochondrial biogenesis using Simple Western assays of term placental villus tissue. We applied consensus clustering, weighted gene correlation network analysis (WGCNA), and principal component analysis (PCA) to derive signaling scores. Model performance (AIC, R\u0026sup2;, and RMSE) was compared, and associations with offspring outcomes at age 4 years (%fat mass; fasting adiponectin, leptin, insulin, glucose, and lipids) were estimated using multivariable linear regression adjusted for offspring age, race and ethnicity, and maternal pre-pregnancy BMI.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eConsensus clustering outperformed PCA and WGCNA based on model fit statistics. The mTOR/AMPK cluster, characterized by activation of mTOR complex 1 and energy sensing (e.g., phosphorylated 4E-BP1, RPS6, AMPK), was inversely associated with childhood %fat mass (β: \u0026minus;\u0026thinsp;2.51%, 95% CI: \u0026minus;\u0026thinsp;4.44, \u0026minus;\u0026thinsp;0.58). The IGF/Mitochondrial Biogenesis cluster was positively associated with childhood triglyceride levels (17.90 [6.14, 29.60] mg/dL).\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e\u003cp\u003eConsensus clustering provided superior model fit compared to WGCNA and PCA. Placental signaling clusters were associated with childhood adiposity and metabolic markers, supporting the relevance of coordinated placental activity to early metabolic programming in a healthy pregnancy cohort. These findings highlight the utility of unsupervised analytic approaches in placental biology and the potential of early-life placental markers to inform pediatric metabolic disease risk. However, which approach is best for summarizing complex protein data is likely dependent on the data structure, dimensionality, and covariance.\u003c/p\u003e","manuscriptTitle":"Unsupervised Approaches to Placental Protein Clustering: Which Best Captures Signals Linked to Childhood Metabolic Health?","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-14 16:07:02","doi":"10.21203/rs.3.rs-7926327/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-12-08T13:49:27+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-07T20:43:32+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-02T22:30:17+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-26T03:03:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"267330343562543492950177956240701595144","date":"2025-11-23T22:33:27+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"309634499532873958025570161550364799514","date":"2025-11-22T00:14:50+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"136512529129437018667050714680869591147","date":"2025-11-11T16:28:18+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-11-04T17:01:34+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-23T11:03:02+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-10-23T09:09:46+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Medicine","date":"2025-10-22T19:36:35+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-medicine","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bmed","sideBox":"Learn more about [BMC Medicine](http://bmcmedicine.biomedcentral.com/)","snPcode":"12916","submissionUrl":"https://submission.nature.com/new-submission/12916/3","title":"BMC Medicine","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"4009d9a7-a8b7-4739-a1e5-26d7b864e9a5","owner":[],"postedDate":"November 14th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-24T12:40:50+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-14 16:07:02","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7926327","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7926327","identity":"rs-7926327","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}