Model Assessment Method Research Articles

Time‐periodic model of wind speed and its application in risk evaluation of wind‐power‐integrated power systems

Risk evaluation of wind-power-integrated power systems requires a distinctive wind speed modelling technique. Existing techniques of mid- and long-term risk assessment mainly focus on annual wind speed pattern rather than its monthly and daily patterns, which cannot reflect the time-varying characteristics of power system risk. This article proposes a time-periodic model of wind speed which incorporates two parts. The first part is the monthly wind speed patterns which can be represented as time-periodic functions, while the other one is the fluctuation in daily wind speeds that can be denoted as a random variable following a certain probability distribution. With this model, a risk evaluation procedure for wind-power-integrated power system is developed. Collected wind speed data from four representative sites in China are used to verify the proposed model. The application of the proposed wind model and risk assessment method is tested by calculating annual and monthly risk indices of a modified IEEE-RTS79 system. Results can provide references for power system planning, mid- and long-term dispatching, and monthly generation scheduling.

Comparative analysis of methods for evaluation of protein models against native structures

Measuring discrepancies between protein models and native structures is at the heart of development of protein structure prediction methods and comparison of their performance. A number of different evaluation methods have been developed; however, their comprehensive and unbiased comparison has not been performed. We carried out a comparative analysis of several popular model assessment methods (RMSD, TM-score, GDT, QCS, CAD-score, LDDT, SphereGrinder and RPF) to reveal their relative strengths and weaknesses. The analysis, performed on a large and diverse model set derived in the course of three latest community-wide CASP experiments (CASP10-12), had two major directions. First, we looked at general differences between the scores by analyzing distribution, correspondence and correlation of their values as well as differences in selecting best models. Second, we examined the score differences taking into account various structural properties of models (stereochemistry, hydrogen bonds, packing of domains and chain fragments, missing residues, protein length and secondary structure). Our results provide a solid basis for an informed selection of the most appropriate score or combination of scores depending on the task at hand. Supplementary data are available at Bioinformatics online.

Asymptomatic hyperuricemia and coronary flow reserve in patients with metabolic syndrome

BackgroundPatients with metabolic syndrome (MetS) are at increased risk of asymptomatic hyperuricemia (i.e., elevated serum uric acid (SUA) level without gout) and cardiovascular disease. We conducted a cross-sectional study to examine associations between SUA levels and coronary flow reserve and urate deposits in carotid arteries in patients with asymptomatic hyperuricemia and MetS.MethodsAdults aged ≥40 years with MetS and SUA levels ≥6.5 mg/dl, but no gout, were eligible. Using a stress myocardial perfusion positron emission tomography (PET), we assessed myocardial blood flow (MBF) at rest and stress and calculated coronary flow reserve (CFR). CFR < 2.0 is considered abnormal and associated with increased cardiovascular risk. We also measured insulin resistance by homeostatic model assessment (HOMA-IR) method and urate deposits using dual-energy CT (DECT) of the neck for the carotid arteries.ResultsForty-four patients with the median age of 63.5 years underwent a blood test, cardiac PET and neck DECT scans. Median (IQR) SUA was 7.8 (7.1–8.4) mg/dL. The median (IQR) CFR was abnormally low at 1.9 (1.7–2.4) and the median (IQR) stress MBF was 1.7 (1.3–2.2) ml/min/g. None had urate deposits in the carotid arteries detected by DECT. In multivariable linear regression analyses, SUA had no association with CFR (β = − 0.12, p = 0.78) or stress MBF (β = − 0.52, p = 0.28). Among non-diabetic patients (n = 25), SUA was not associated with HOMA-IR (β = 2.08, p = 0.10).ConclusionsAmong MetS patients with asymptomatic hyperuricemia, we found no relationship between SUA and CFR, stress MBF, and insulin resistance. No patients had any DECT detectable subclinical urate deposition in the carotid arteries.

MACS – a new SPM toolbox for model assessment, comparison and selection

BackgroundIn cognitive neuroscience, functional magnetic resonance imaging (fMRI) data are widely analyzed using general linear models (GLMs). However, model quality of GLMs for fMRI is rarely assessed, in part due to the lack of formal measures for statistical model inference. New methodWe introduce a new SPM toolbox for model assessment, comparison and selection (MACS) of GLMs applied to fMRI data. MACS includes classical, information-theoretic and Bayesian methods of model assessment previously applied to GLMs for fMRI as well as recent methodological developments of model selection and model averaging in fMRI data analysis. ResultsThe toolbox – which is freely available from GitHub – directly builds on the Statistical Parametric Mapping (SPM) software package and is easy-to-use, general-purpose, modular, readable and extendable. We validate the toolbox by reproducing model selection and model averaging results from earlier publications. Comparison with existing methodsA previous toolbox for model diagnosis in fMRI has been discontinued and other approaches to model comparison between GLMs have not been translated into reusable computational resources in the past. ConclusionsIncreased attention on model quality will lead to lower false-positive rates in cognitive neuroscience and increased application of the MACS toolbox will increase the reproducibility of GLM analyses and is likely to increase the replicability of fMRI studies.

The relationship between insulin resistance/β-cell dysfunction and diabetic retinopathy in Chinese patients with type 2 diabetes mellitus: the Desheng Diabetic Eye Study

To investigate the relationship between insulin resistance (IR)/β-cell dysfunction and diabetic retinopathy (DR) in Chinese patients with type 2 diabetes mellitus (T2DM), and to explore further whether there were differences in the relationship among diabetic patients with higher and lower body mass index (BMI). Cross-sectional study. A total of 1466 subjects with T2DM were recruited in a local Desheng Community of urban Beijing from November 2009 to June 2012 for the cohort of Beijing Desheng Diabetic Eye Study. Standardized evaluation was carried out for each participant, including questionnaire, ocular and anthropometric examinations, and laboratory tests. Seven fields 30° color fundus photographs were used for DR grading according to the Early Treatment Diabetic Retinopathy Study protocols. Homeostatis Model Assessment (HOMA) method was employed for IR and β-cell function assessment. After excluding those participants who were treated with insulin (n=352) or had missing data of fasting insulin (n=96), and further excluding those with poor quality of retinal photographs (n=10), a total of 1008 subjects were included for the final analysis, 406 (40.3%) were men and 602 (59.7%) were women, age ranging from 34 to 86 (64.87±8.28)y. Any DR (levels 14 and above) was present in 278 (27.6%) subjects. After adjusting for possible covariates, the presence of any DR did not correlate with HOMA IR [odds ratio (OR) 1.51, 95% confidence interval (CI) 0.87-2.61, P=0.14] or HOMA β-cell (OR 0.71, 95%CI 0.40-1.26, P=0.25). After stratification by BMI, the presence of any DR was associated positively with HOMA IR (OR 2.46, 95%CI: 1.18-5.12, P=0.016), and negatively with HOMA β-cell (OR 0.40, 95%CI: 0.19-0.87, P=0.021) in the group of patients with higher BMI (≥25 kg/m2). In the group of patients with lower BMI (<25 kg/m2), the presence of any DR was not associated with HOMA IR (OR 1.00, 95%CI: 0.43-2.33, P=1.00) or HOMA β-cell (OR 1.41, 95%CI: 0.60-3.32, P=0.43). The data suggest that higher IR and lower β-cell function are associated with the presence of DR in the subgroup of diabetic patients with higher BMI. However, this association is not statistically significant in diabetic patients with lower BMI.

The effect of Tehran metro rail system on residential property values: A comparative analysis between high-income and low-income neighbourhoods

Estimating the effect of rail transit on residential property values has resulted in mixed findings. Some researchers report positive effects on property values while others are negative or insignificant. The current paper argues that the impact of rail transit on property values depends on contextual factors which can influence magnitude and direction of the impact and cause variation in the findings. Tehran’s Metro Rail System is chosen because the neighbourhoods served by the metro to the north side and south side of the city are dramatically different in terms of economic, social and physical circumstances. A comparative analysis of six of Tehran’s metro stations is conducted between the high-income neighbourhoods, which are located primarily on the city’s north side, and lower-income neighbourhoods on the south side to determine the effects of proximity to metro stations on residential property values. The paper uses a mixed-method sequential explanatory design based on a before-and-after estimation strategy, which includes trend analysis, difference-in-differences model and qualitative impact assessment methods. The results show that there are large increases in premiums for residential properties near the lower-income neighbourhoods, the south side parts of the Tehran Metro Rail System, while a negative treatment effect for residential properties lying close to the northern stations in the high-income neighbourhoods. The qualitative survey also suggests that the impact of metro station is affected by a number of contextual factors, including the need for public transportation, land-use planning and management, socio-cultural effect and possible nuisance effects.

PhyloMAd: efficient assessment of phylogenomic model adequacy.

Statistical phylogenetic inference plays an important role in evolutionary biology. The accuracy of phylogenetic methods relies on having suitable models of the evolutionary process. Various tools allow comparisons of candidate phylogenetic models, but assessing the absolute performance of models remains a considerable challenge. We introduce PhyloMAd, a user-friendly application for assessing the adequacy of commonly used models of nucleotide substitution and among-lineage rate variation. Our software implements a fast, likelihood-based method of model assessment that is tractable for analyses of large multi-locus datasets. PhyloMAd provides a means of informing model improvement, or selecting data to enhance the evolutionary signal in phylogenomic analyses. PhyloMAd, together with a manual, a tutorial and the source code, are freely available from the GitHub repository github.com/duchene/phylomad.

Slice–Gibbs sampling algorithm for estimating the parameters of a multilevel item response model

In a fully Bayesian framework, a novel slice–Gibbs algorithm is developed to estimate a multilevel item response theory (IRT) model. The advantage of this algorithm is that it can recover parameters well based on various types of prior distributions of the item parameters, including informative and non-informative priors. In contrast to the traditional Metropolis–Hastings (M–H) within Gibbs algorithm, the slice–Gibbs algorithm is faster and more efficient, due to its drawing the sample with acceptance probability as one, rather than tuning the proposal distributions to achieve the reasonable acceptance probabilities, especially for the logistic model without conjugate distribution. In addition, based on the Markov chain Monte Carlo (MCMC) output, two model assessment methods are investigated concerning the goodness of fit between models. The information criterion method on the basis of marginal likelihood is proposed to assess the different structural multilevel models, and the cross-validation method is used to evaluate the overall multilevel IRT models. The feasibility and effectiveness of the slice–Gibbs algorithm are investigated in simulation studies. An application using a real data involving students’ mathematics test achievements is reported.

Numerical simulation and prediction of groundwater environmental contamination in a coastal polyurethane plant based on GMS

GMS was used to simulate the groundwater pollution status during the operation period of a coastal polyurethane factory. Based on the geological and hydraulic data, the groundwater flow field model and solute transport model were established using the MODFLOW and MT3DMS module respectively by which the potential area and concentration of pollutant were predicted and the extent of groundwater pollution in the study area was evaluated. The simulation and prediction results showed that under abnormal conditions, the groundwater environment would be affected. As the characteristic pollutants, COD and ethylene glycol mainly migrated along the direction of water flow and exceeded the standard for a certain period of time. The migration distance and pollution range of contaminants expanded with time, and the peak concentration decreased with the increase of diffusion distance. After 50 years, the potential pollution area of COD and ethylene glycol would reach 225651.2m2 and 210947.5m2 respectively. During the migration process, the concentration of pollutants in the groundwater gradually recovered to the background value, and the over-standard area was reduced to zero. The comprehensive prediction model and risk assessment method of this study can help to formulate practical prevention and control measures for the protection of groundwater environment.

Vitamin K2 alleviates type 2 diabetes in rats by induction of osteocalcin gene expression.

The biological mechanisms behind the association between vitamin K (Vit K) and glucose metabolism are uncertain. We aimed to analyze the expression of insulin 1 (Ins 1), insulin 2 (Ins 2) and cyclin D2, the expression of adiponectin and UCP-1 . In addition, we aimed to estimate the doses of Vit K2 able to affect various aspects of glucose and energy metabolism in type 2 diabetes. Thirty adult male rats were allocated equally into five groups: control group, diabetes mellitus group, and groups 3, 4, and 5, which received Vit K2 at three daily dose levels (10, 15, and 30mg/kg, respectively) for 8wk. At the end of the study, blood samples were collected to quantify total osteocalcin, fasting plasma glucose, fasting insulin, and relevant variables. The expression of OC, Ins 1, Ins 2, cyclin D2, adiponectin, UCP-1 genes was analyzed by real-time polymerase chain reaction. After administration of Vit K2, a dose-dependent decrease in fasting plasma glucose, hemoglobin A1c and homeostatic model assessment method insulin resistance, and a dose-dependent increase in fasting insulin and homeostatic model assessment method β cell function levels, when compared with diabetes mellitus rats, were detected. There was significant upregulation of OC, Ins 1, Ins 2, or cyclin D2 gene expression in the three treated groups in a dose-dependent manner when compared with the diabetic rats. However, expression of adiponectin and UCP-1 were significantly increased at the highest dose (30mg/kg daily) only. Vit K2 administration could improve glycemic status in type 2 diabetic rats by induction of OC gene expression. Osteocalcin could increase β-cell proliferation, energy expenditure, and adiponectin expression. Different concentrations of Vit K2 were required to affect glucose metabolism and insulin sensitivity.

A New Method for Logistic Model Assessment

It is well known that the logistic model plays an important role for the analysis of binary outcomes. Most of the existing methods for the assessment of logistic models are constructed based on the distance between the observed and the predicted outcomes. We consider a new method from a different perspective by assessing the distance between two consistent estimators developed under the same logistic model form. The proposed tests are easy to implement and are applicable to both prospective and case-control studies.

The outcomes of glucose abnormalities in chronic hepatitis C patients receiving interferon-free direct antiviral agents

Direct-acting antiviral agents (DAAs) have been widely used for chronic hepatitis C (CHC) treatment recently. The characteristics of glucose abnormalities after DAAs therapy however, remain elusive. We aimed to elucidate the mutual impact between treatment response and parameters of glucose abnormalities after DAAs therapy in CHC patients. CHC patients who received DAAs therapy were recruited. The primary outcome measurements were their insulin resistance (IR) and beta-cell function assessed by the homeostasis model assessment (HOMA) method before treatment and at end-of-follow-up (EOF). Sixty-five CHC patients (19 males, mean age = 59.8 ± 10.3 years) were consecutively enrolled. They included 47 (72.3%) patients of genotype-1 infection. The treatment regimens among patients were sofosbuvir in 30 patients, paritaprevir-ritonavir/ombitasvir/dasabuvir in 23 patients, and asunaprevir/daclatasvir in 12 patients respectively. The overall sustained virological response rate was 98.5%. The mean IR at EOF was 2.6 ± 1.8, which was not significantly different from baseline level (2.7 ± 2.9, P = 0.75). There was a significant improvement of beta-cell function at EOF compared to baseline (107.7 ± 86.8 to 86.7 ± 44.5, P = 0.05). The amelioration of beta-cell function at EOF was significantly observed among 23 patients of high baseline IR (166.7 ± 111.3 of baseline vs 105.7 ± 48.2 of EOF, P = 0.04). Six (60%) of the 10 pre-diabetic patients at baseline achieved a normoglycemic state at EOF. Successful eradication of HCV by DAAs might improve glucose abnormalities in CHC patients, particularly among those who had high IR.

Model assessment in scientific computing

PurposeThis paper aims to focus on the assessment of the ability of computer models with imperfect functional forms and uncertain input parameters to represent reality.Design/methodology/approachIn this assessment, both the agreement between a model’s predictions and available experiments and the robustness of this agreement to uncertainty have been evaluated. The concept of satisfying boundaries to represent input parameter sets that yield model predictions with acceptable fidelity to observed experiments has been introduced.FindingsSatisfying boundaries provide several useful indicators for model assessment, and when calculated for varying fidelity thresholds and input parameter uncertainties, reveal the trade-off between the robustness to uncertainty in model parameters, the threshold for satisfactory fidelity and the probability of satisfying the given fidelity threshold. Using a controlled case-study example, important modeling decisions such as acceptable level of uncertainty, fidelity requirements and resource allocation for additional experiments are shown.Originality/valueTraditional methods of model assessment are solely based on fidelity to experiments, leading to a single parameter set that is considered fidelity-optimal, which essentially represents the values which yield the optimal compensation between various sources of errors and uncertainties. Rather than maximizing fidelity, this study advocates for basing model assessment on the model’s ability to satisfy a required fidelity (or error tolerance). Evaluating the trade-off between error tolerance, parameter uncertainty and probability of satisfying this predefined error threshold provides us with a powerful tool for model assessment and resource allocation.

Risk factors for Complex and Severe Coronary Artery Disease in Type 2 Diabetes Mellitus

Background: Type 2 diabetes mellitus is often associated with severe Coronary artery disease (CAD). Since patients with higher risk of severe disease are likely to get better benefit from aggressive management, it is essential to identify factors which are associated with severe macrovascular disease. We looked at the possibility of hyperinsulinemia being a marker for severe and complex coronary artery disease in type 2 diabetes mellitus, to select patients who would benefit from aggressive treatment. Methods: A cross sectional study of 290 type 2 diabetic patients, who underwent coronary angiogram for the evaluation of clinically suspected CAD at a tertiary care hospital were recruited. Biochemical and anthropometric parameters were analysed. Insulin resistance was measured by homeostasis model assessment method. Angiographically measured syntax score of more than 22 is considered to be severe and complex CAD. Receiver operating curve characteristic was performed to find out the optimal cut-off value for insulin resistance and fasting insulin. Predictors of syntax score greater than 22 were identified by multiple logistic regression analysis. Results: An insulin level > 20 μIU/ml (OR: 6.86, 95% CI: 2.25-20.88) emerged as an independent predictor of severe and complex CAD. The optimal cut-off of insulin for predicting severe CAD was 20 with sensitivity and specificity of 80% (95% CI: 0.68 - 0.88) and 79% (95% CI: 0.73 - 0.83) respectively. Conclusion: Hyperinsulinemia could serve as a marker to identify severe and complex CAD in type 2 diabetes at an earlier stage of diabetes. Key words: Coronary Artery Disease, Hyperinsulinemia, Insulin resistance, Syntax score, Type 2 diabetes mellitus.

A clinical study of insulin resistance in patients with chronic pancreatitis

ObjectiveInsulin resistance (IR) and beta cell dysfunction are the pathophysiological determinants of the diabetes mellitus (DM). We investigated the presence of IR in patients with chronic pancreatitis (CP) and compared the same with the underlying etiology. MethodsIn this cross-sectional, observational study, we included serial patients of CP presented to our hospital. The study population is in different stages of CP and are grouped as alcoholic CP (Group 1; N=67) and tropical CP (Group 2; N=35). IR was estimated by the homeostasis model assessment (HOMA) method. The results were analyzed by appropriate statistical methods. ResultsThe study participants (85M and 17F) had a mean age 40.8±12.6 yr, CP duration 3.7±4.7 yr and body mass index (BMI) of 22.5±3.2kg/m2. DM was seen in 54 patients with average glycosylated hemoglobin of 7.5±1.6%. A total of 9 patients had HOMA-IR more than 3 suggestive of IR with no significant difference between the two groups. The duration of the DM correlated negatively with glycemic parameters and BMI showed a positive correlation with the fasting insulin and HOMA-IR. ConclusionIR was seen in a minority of patients with CP and is not a significant contributor to the pancreatogenic diabetes.

Data-Driven Risk Assessment from Small Scale Epidemics: Estimation and Model Choice for Spatio-Temporal Data with Application to a Classical Swine Fever Outbreak.

Livestock epidemics have the potential to give rise to significant economic, welfare, and social costs. Incursions of emerging and re-emerging pathogens may lead to small and repeated outbreaks. Analysis of the resulting data is statistically challenging but can inform disease preparedness reducing potential future losses. We present a framework for spatial risk assessment of disease incursions based on data from small localized historic outbreaks. We focus on between-farm spread of livestock pathogens and illustrate our methods by application to data on the small outbreak of Classical Swine Fever (CSF) that occurred in 2000 in East Anglia, UK. We apply models based on continuous time semi-Markov processes, using data-augmentation Markov Chain Monte Carlo techniques within a Bayesian framework to infer disease dynamics and detection from incompletely observed outbreaks. The spatial transmission kernel describing pathogen spread between farms, and the distribution of times between infection and detection, is estimated alongside unobserved exposure times. Our results demonstrate inference is reliable even for relatively small outbreaks when the data-generating model is known. However, associated risk assessments depend strongly on the form of the fitted transmission kernel. Therefore, for real applications, methods are needed to select the most appropriate model in light of the data. We assess standard Deviance Information Criteria (DIC) model selection tools and recently introduced latent residual methods of model assessment, in selecting the functional form of the spatial transmission kernel. These methods are applied to the CSF data, and tested in simulated scenarios which represent field data, but assume the data generation mechanism is known. Analysis of simulated scenarios shows that latent residual methods enable reliable selection of the transmission kernel even for small outbreaks whereas the DIC is less reliable. Moreover, compared with DIC, model choice based on latent residual assessment correlated better with predicted risk.

Concentrations of leptin, adiponectin and other metabolic parameters in non-obese children with Down syndrome.

Recent data indicates that adults with Down syndrome (DS) are at increased risk for cardiovascular disease (CVD) that significantly contributes to their morbidity and mortality. Although identification of cardiometabolic risk factors during childhood is desirable to design preventive interventions, the data on such risk factors in children with DS is scarce. The aim of this study was to study the cardiometabolic risk factors such as insulin resistance (IR), leptin and adiponectin concentrations, lipid abnormalities and leptin resistance in non-obese children with DS. This cross-sectional case control study included karyotype confirmed trisomy-21 DS children aged 2-12 years and their matched healthy controls. After detailed anthropometry, weight, height and body mass index (BMI) standard deviation scores (SDSs) were calculated with reference data. Laboratory evaluation included determination of fasting lipid parameters, insulin, glucose, leptin and adiponectin concentrations. The homeostasis model assessment method (HOMA-IR) was used to assess IR and the ratio of leptin to BMI was used as an index of leptin resistance. Seventy-seven children (39 with DS and 38 controls) comprised the study cohort. The anthropometric parameters were similar in the two groups. Children with DS showed significantly higher mean leptin concentrations (2.098±1.68 ng/mL vs. 1.44±0.52 ng/mL, p-value: 0.00) and higher indices of leptin resistance (0.127±0.085 vs. 0.09±0.03, p-value: 0.001) as compared to controls. Fasting adiponectin concentrations were lower (20.64±19.87 ng/mL vs. 32.58±34.25 ng/mL, p-value: 0.21) and fasting glucose higher (89.25±8.12 mg/dL vs. 85.71±5.52 mg/dL, p-value: 0.06) in the DS group as compared to the controls but the differences did not reach statistical significance. The concentrations of insulin, various lipid parameters and calculated HOMA-IR values were similar in the two groups. In the DS group, five children were identified to have high (>75th centile) leptin levels and four as impaired fasting glucose as compared to none in the controls. Alterations of several cardiometabolic risk factors, in particular, leptin concentrations and leptin resistance are present in children with DS. The presence of hyperleptinemia without hyperinsulinemia suggests a probable inherent genetic basis for increased leptin resistance in patients with DS. There is a need for larger studies to further understand increased leptin resistance in DS that may contribute to increased CVD related morbidity and mortality in these patients.

Tscount: An R Package for Analysis of Count Time Series Following Generalized Linear Models

The R package tscount provides likelihood-based estimation methods for analysis and modeling of count time series following generalized linear models. This is a flexible class of models which can describe serial correlation in a parsimonious way. The conditional mean of the process is linked to its past values, to past observations and to potential covariate effects. The package allows for models with the identity and with the logarithmic link function. The conditional distribution can be Poisson or negative binomial. An important special case of this class is the so-called INGARCH model and its log-linear extension. The package includes methods for model fitting and assessment, prediction and intervention analysis. This paper summarizes the theoretical background of these methods. It gives details on the implementation of the package and provides simulation results for models which have not been studied theoretically before. The usage of the package is illustrated by two data examples. Additionally, we provide a review of R packages which can be used for count time series analysis. This includes a detailed comparison of tscount to those packages.

Effect of Insulin Resistance in Obese Polycystic Ovarian Disease

The study was mainly focussed to measure the insulin resistance by HOMAIR values (homeostasis model assessment method: insulin resistance) and found that in all patients with obese Polycystic ovarian syndrome (PCOS) showed increased insulin resistance compared to control subjects with normal body weight and ovulatory menstruation.

Osteocalcin: A New Biomarker for Non Alcoholic Fatty Liver Disease (NAFLD) in Children and Adolescents

Background: Nonalcoholic fatty liver disease (NAFLD) is a common liver disease in adults but uncommon in Pediatrics. Patients with chronic liver disease are at increasing risk of developing metabolic bone disease which may be attributed to decreased Osteocalcin which is involved in bone extracellular matrix mineralization. Aim of the work: was to measure serum osteocalcin levels in children and adolescents with NAFLD and to investigate the relationship with variables degrees of NAFLD. Materials and methods: This study was carried out upon 60 children with NAFL detected by abdominal ultrasonography. Forty apparently healthy children, matched for age, sex and body mass index (BMI) were chosen as controls. They were subjected to history taking, clinical examination, investigations included liver function tests, fasting serum glucose and insulin levels, Homeostasis Model Assessment method of Insulin Resistance (HOMA-IR), serum osteocalcin level and abdominal Ultrasonography to assess semiquantitavely the degree of steatosis. Results: Mean serum osteocalcin level was significantly lower in patients than in controls. Mean serum level of ALT, total cholesterol and triglycerides, fasting insulin and HOMA-IR were markedly increases as the hepatic steatosis advancing in severity. There was a parallel decrease of mean serum osteocalcin level with advancing the degree of hepatic steatosis by ultrasound. Serum osteocalcin levels were inversely correlated with the values of W/H ratio, fasting insulin and HOMA-IR. Conclusions: NAFLD had lower serum levels of osteocalcin compared to controls. It were inversely associated with the degrees of steatosis so can considered as a biomarker for severity NAFLD in pediatric age.