MRSL: a causal network pruning algorithm based on GWAS summary data.

Causal discovery is a powerful tool to disclose underlying structures by analyzing purely observational data. Genetic variants can provide useful complementary information for structure learning. Recently, Mendelian randomization (MR) studies have provided abundant marginal causal relationships of traits. Here, we propose a causal network pruning algorithm MRSL (MR-based structure learning algorithm) based on these marginal causal relationships. MRSL combines the graph theory with multivariable MR to learn the conditional causal structure using only genome-wide association analyses (GWAS) summary statistics. Specifically, MRSL utilizes topological sorting to improve the precision of structure learning. It proposes MR-separation instead of d-separation and three candidates of sufficient separating set for MR-separation. The results of simulations revealed that MRSL had up to 2-fold higher F1 score and 100 times faster computing time than other eight competitive methods. Furthermore, we applied MRSL to 26 biomarkers and 44 International Classification of Diseases 10 (ICD10)-defined diseases using GWAS summary data from UK Biobank. The results cover most of the expected causal links that have biological interpretations and several new links supported by clinical case reports or previous observational literatures.

Impact of nonrandom selection mechanisms on the causal effect estimation for two-sample Mendelian randomization methods.

Nonrandom selection in one-sample Mendelian Randomization (MR) results in biased estimates and inflated type I error rates only when the selection effects are sufficiently large. In two-sample MR, the different selection mechanisms in two samples may more seriously affect the causal effect estimation. Firstly, we propose sufficient conditions for causal effect invariance under different selection mechanisms using two-sample MR methods. In the simulation study, we consider 49 possible selection mechanisms in two-sample MR, which depend on genetic variants (G), exposures (X), outcomes (Y) and their combination. We further compare eight pleiotropy-robust methods under different selection mechanisms. Results of simulation reveal that nonrandom selection in sample II has a larger influence on biases and type I error rates than those in sample I. Furthermore, selections depending on X+Y, G+Y, or G+X+Y in sample II lead to larger biases than other selection mechanisms. Notably, when selection depends on Y, bias of causal estimation for non-zero causal effect is larger than that for null causal effect. Especially, the mode based estimate has the largest standard errors among the eight methods. In the absence of pleiotropy, selections depending on Y or G in sample II show nearly unbiased causal effect estimations when the casual effect is null. In the scenarios of balanced pleiotropy, all eight MR methods, especially MR-Egger, demonstrate large biases because the nonrandom selections result in the violation of the Instrument Strength Independent of Direct Effect (InSIDE) assumption. When directional pleiotropy exists, nonrandom selections have a severe impact on the eight MR methods. Application demonstrates that the nonrandom selection in sample II (coronary heart disease patients) can magnify the causal effect estimation of obesity on HbA1c levels. In conclusion, nonrandom selection in two-sample MR exacerbates the bias of causal effect estimation for pleiotropy-robust MR methods.

Isotopic evidence for increased carbon and nitrogen exchanges between peatland plants and their symbiotic microbes with rising atmospheric CO2 concentrations since 15,000 cal. year BP.

Whether nitrogen (N) availability will limit plant growth and removal of atmospheric CO2 by the terrestrial biosphere this century is controversial. Studies have suggested that N could progressively limit plant growth, as trees and soils accumulate N in slowly cycling biomass pools in response to increases in carbon sequestration. However, a question remains over whether longer-term (decadal to century) feedbacks between climate, CO2 and plant N uptake could emerge to reduce ecosystem-level N limitations. The symbioses between plants and microbes can help plants to acquire N from the soil or from the atmosphere via biological N2 fixation-the pathway through which N can be rapidly brought into ecosystems and thereby partially or completely alleviate N limitation on plant productivity. Here we present measurements of plant N isotope composition (δ15 N) in a peat core that dates to 15,000 cal. year BP to ascertain ecosystem-level N cycling responses to rising atmospheric CO2 concentrations. We find that pre-industrial increases in global atmospheric CO2 concentrations corresponded with a decrease in the δ15 N of both Sphagnum moss and Ericaceae when constrained for climatic factors. A modern experiment demonstrates that the δ15 N of Sphagnum decreases with increasing N2 -fixation rates. These findings suggest that plant-microbe symbioses that facilitate N acquisition are, over the long term, enhanced under rising atmospheric CO2 concentrations, highlighting an ecosystem-level feedback mechanism whereby N constraints on terrestrial carbon storage can be overcome.

Causal associations between female reproductive behaviors and psychiatric disorders: a lifecourse Mendelian randomization study.

BACKGROUND: The timings of reproductive life events have been examined to be associated with various psychiatric disorders. However, studies have not considered the causal pathways from reproductive behaviors to different psychiatric disorders. This study aimed to investigate the nature of the relationships between five reproductive behaviors and twelve psychiatric disorders. METHODS: Firstly, we calculated genetic correlations between reproductive factors and psychiatric disorders. Then two-sample Mendelian randomization (MR) was conducted to estimate the causal associations among five reproductive behaviors, and these reproductive behaviors on twelve psychiatric disorders, using genome-wide association study (GWAS) summary data from genetic consortia. Multivariable MR was then applied to evaluate the direct effect of reproductive behaviors on these psychiatric disorders whilst accounting for other reproductive factors at different life periods. RESULTS: Univariable MR analyses provide evidence that age at menarche, age at first sexual intercourse and age at first birth have effects on one (depression), seven (anxiety disorder, ADHD, bipolar disorder, bipolar disorder II, depression, PTSD and schizophrenia) and three psychiatric disorders (ADHD, depression and PTSD) (based on p<7.14×10-4), respectively. However, after performing multivariable MR, only age at first sexual intercourse has direct effects on five psychiatric disorders (Depression, Attention deficit or hyperactivity disorder, Bipolar disorder, Posttraumatic stress disorder and schizophrenia) when accounting for other reproductive behaviors with significant effects in univariable analyses. CONCLUSION: Our findings suggest that reproductive behaviors predominantly exert their detrimental effects on psychiatric disorders and age at first sexual intercourse has direct effects on psychiatric disorders.

Quantifying substantial carcinogenesis of genetic and environmental factors from measurement error in the number of stem cell divisions.

BACKGROUND: The relative contributions of genetic and environmental factors versus unavoidable stochastic risk factors to the variation in cancer risk among tissues have become a widely-discussed topic. Some claim that the stochastic effects of DNA replication are mainly responsible, others believe that cancer risk is heavily affected by environmental and hereditary factors. Some of these studies made evidence from the correlation analysis between the lifetime number of stem cell divisions within each tissue and tissue-specific lifetime cancer risk. However, they did not consider the measurement error in the estimated number of stem cell divisions, which is caused by the exposure to different levels of genetic and environmental factors. This will obscure the authentic contribution of environmental or inherited factors. METHODS: In this study, we proposed two distinct modeling strategies, which integrate the measurement error model with the prevailing model of carcinogenesis to quantitatively evaluate the contribution of hereditary and environmental factors to cancer development. Then, we applied the proposed strategies to cancer data from 423 registries in 68 different countries (global-wide), 125 registries across China (national-wide of China), and 139 counties in Shandong province (Shandong provincial, China), respectively. RESULTS: The results suggest that the contribution of genetic and environmental factors is at least 92% to the variation in cancer risk among 17 tissues. Moreover, mutations occurring in progenitor cells and differentiated cells are less likely to be accumulated enough for cancer to occur, and the carcinogenesis is more likely to originate from stem cells. Except for medulloblastoma, the contribution of genetic and environmental factors to the risk of other 16 organ-specific cancers are all more than 60%. CONCLUSIONS: This work provides additional evidence that genetic and environmental factors play leading roles in cancer development. Therefore, the identification of modifiable environmental and hereditary risk factors for each cancer is highly recommended, and primary prevention in early life-course should be the major focus of cancer prevention.

A causal data fusion method for the general exposure and outcome.

With the advent of the big data era, the need to combine multiple individual data sets to draw causal effects arises naturally in many medical and biological applications. Especially each data set cannot measure enough confounders to infer the causal effect of an exposure on an outcome. In this article, we extend the method proposed by a previous study to causal data fusion of more than two data sets without external validation and to a more general (continuous or discrete) exposure and outcome. Theoretically, we obtain the condition for identifiability of exposure effects using multiple individual data sources for the continuous or discrete exposure and outcome. The simulation results show that our proposed causal data fusion method has unbiased causal effect estimate and higher precision than traditional regression, meta-analysis and statistical matching methods. We further apply our method to study the causal effect of BMI on glucose level in individuals with diabetes by combining two data sets. Our method is essential for causal data fusion and provides important insights into the ongoing discourse on the empirical analysis of merging multiple individual data sources.

Exploring the Causal Roles of Circulating Remnant Lipid Profile on Cardiovascular and Cerebrovascular Diseases: Mendelian Randomization Study.

BACKGROUND: Causal evidence of circulating lipids especially the remnant cholesterol with cardiovascular and cerebrovascular disease (CVD) is lacking. This research aimed to explore the causal roles of extensive lipid traits especially the remnant lipids in CVD. METHODS: Two-sample Mendelian randomization (TSMR) analysis was performed based on large-scale meta-analysis datasets in European ancestry. The causal effect of 15 circulating lipid profiles including 6 conventional lipids and 9 remnant lipids on coronary heart disease (CHD) and ischemic stroke (IS), as well as the subtypes, was assessed. RESULTS: Apolipoprotein B (Apo B), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and triglyceride (TG) were still important risk factors for CHD and myocardial infarction (MI) but not for IS. Apo B is the strongest which increased the CHD and MI risk by 44% and 41%, respectively. The odds ratios (ORs) of total TG on CHD and MI were 1.25 (95% confidence interval [CI], 1.13-1.38) and 1.24 (95% CI, 1.11-1.38), respectively. A one standard deviation difference increased TG in medium very-low-density lipoproteins (M.VLDL.TG), TG in small VLDL (S.VLDL.TG), TG in very small VLDL (XS.VLDL.TG), TG in intermediate-density lipoproteins (IDL.TG), TG in very large HDL (XL.HDL.TG), and TG in small HDL (S.HDL.TG) particles also robustly increased the risk of CHD and MI by 9-28% and 9-27%, respectively. TG in very/extremely large VLDL (XXL.VLDL.TG and XL.VLDL.TG) were insignificant or even negatively associated with CHD (in multivariable TSMR), and negatively associated with IS as well. CONCLUSION: The remnant lipids presented heterogeneity and two-sided effects for the risk of CHD and IS that may partially rely on the particle size. The findings suggested that the remnant lipids were required to be intervened according to specific components. This research confirms the importance of remnant lipids and provides causal evidence for potential targets for intervention.

Identification and Estimation of Causal Effects Using a Negative-Control Exposure in Time-Series Studies With Applications to Environmental Epidemiology.

The initial aim of environmental epidemiology is to estimate the causal effects of environmental exposures on health outcomes. However, due to lack of enough covariates in most environmental data sets, current methods without enough adjustments for confounders inevitably lead to residual confounding. We propose a negative-control exposure based on a time-series studies (NCE-TS) model to effectively eliminate unobserved confounders using an after-outcome exposure as a negative-control exposure. We show that the causal effect is identifiable and can be estimated by the NCE-TS for continuous and categorical outcomes. Simulation studies indicate unbiased estimation by the NCE-TS model. The potential of NCE-TS is illustrated by 2 challenging applications: We found that living in areas with higher levels of surrounding greenness over 6 months was associated with less risk of stroke-specific mortality, based on the Shandong Ecological Health Cohort during January 1, 2010, to December 31, 2018. In addition, we found that the widely established negative association between temperature and cancer risks was actually caused by numbers of unobserved confounders, according to the Global Open Database from 2003-2012. The proposed NCE-TS model is implemented in an R package (R Foundation for Statistical Computing, Vienna, Austria) called NCETS, freely available on GitHub.

Identifying causality, genetic correlation, priority and pathways of large-scale complex exposures of breast and ovarian cancers.

BACKGROUND: Genetic correlations, causalities and pathways between large-scale complex exposures and ovarian and breast cancers need systematic exploration. METHODS: Mendelian randomisation (MR) and genetic correlation (GC) were used to identify causal biomarkers from 95 cancer-related exposures for risk of breast cancer [BC: oestrogen receptor-positive (ER + BC) and oestrogen receptor-negative (ER - BC) subtypes] and ovarian cancer [OC: high-grade serous (HGSOC), low-grade serous, invasive mucinous (IMOC), endometrioid (EOC) and clear cell (CCOC) subtypes]. RESULTS: Of 31 identified robust risk factors, 16 were new causal biomarkers for BC and OC. Body mass index (BMI), body fat mass (BFM), comparative body size at age 10 (CBS-10), waist circumference (WC) and education attainment were shared risk factors for overall BC and OC. Childhood obesity, BMI, CBS-10, WC, schizophrenia and age at menopause were significantly associated with ER + BC and ER - BC. Omega-6:omega-3 fatty acids, body fat-free mass and basal metabolic rate were positively associated with CCOC and EOC; BFM, linoleic acid, omega-6 fatty acids, CBS-10 and birth weight were significantly associated with IMOC; and body fat percentage, BFM and adiponectin were significantly associated with HGSOC. Both GC and MR identified 13 shared factors. Factors were stratified into five priority levels, and visual causal networks were constructed for future interventions. CONCLUSIONS: With analysis of large-scale exposures for breast and ovarian cancers, causalities, genetic correlations, shared or specific factors, risk factor priority and causal pathways and networks were identified.

Morinda officinalis oligosaccharides ameliorate depressive-like behaviors in poststroke rats through upregulating GLUT3 to improve synaptic activity.

Poststroke depression (PSD) is one of the most common psychiatric diseases afflicting stroke survivors, yet the underlying mechanism is poorly understood. The pathophysiology of PSD is presumably multifactorial, involving ischemia-induced disturbance in the context of psychosocial distress. The homeostasis of glucose metabolism is crucial to neural activity. In this study, we showed that glucose consumption was decreased in the medial prefrontal cortex (mPFC) of PSD rats. The suppressed glucose metabolism was due to decreased glucose transporter-3 (GLUT3) expression, the most abundant and specific glucose transporter of neurons. We also found Morinda officinalis oligosaccharides (MOOs), approved as an antidepressive Chinese medicine, through upregulating GLUT3 expression in the mPFC, improved glucose metabolism, and enhanced synaptic activity, which ultimately ameliorated depressive-like behavior in PSD rats. We further confirmed the mechanism that MOOs induce GLUT3 expression via the PKA/pCREB pathway in PSD rats. Our work showed that MOOs treatment is capable of restoring GLUT3 level to improve depressive-like behaviors in PSD rats. We also propose GLUT3 as a potential therapeutic target for PSD and emphasize the importance of metabolism disturbance in PSD pathology.

Exploring the causal pathway from ischemic stroke to atrial fibrillation: a network Mendelian randomization study.

BACKGROUND AND PURPOSE: Previous studies have found ischemic stroke is associated with atrial fibrillation. However, the causal association between ischemic stroke and atrial fibrillation is not clear. Furthermore, the network relationship among ischemic stroke, atrial fibrillation and its risk factors need further attention. This study aims to examine the potential causal association between ischemic stroke and atrial fibrillation and further to explore potential mediators in the causal pathway from ischemic stroke to atrial fibrillation. METHODS: Summary statistics from the ISGC (case = 10,307 and control = 19,326) were used as ischemic stroke genetic instruments, AFGen Consortium data (case = 65,446 and control = 522,744) were used for atrial fibrillation, and other consortia data were used for potential mediators (fasting insulin, white blood cell count, procalcitonin, systolic and diastolic blood pressure, body mass index, waist circumference, and height). Under the framework of network Mendelian randomization, two-sample Mendelian randomization study was performed using summary statistics from several genome-wide association studies. Inverse-variance weighted method was performed to estimate causal effect. RESULTS: Blood pressure mediates the causal pathways from ischemic stroke to atrial fibrillation. The total odds ratio of ischemic stroke on atrial fibrillation was 1.05 (95% confidence interval [CI], 1.02 to 1.07; P = 1.3 × 10-5). One-unit increase of genetically determined ischemic stroke was associated with 0.02 (DBP: 95% CI, 0.001 to 0.034, P = 0.029; SBP: 95% CI, 0.006 to 0.034, P = 0.003) upper systolic and diastolic blood pressure levels. Higher genetically determined systolic and diastolic blood pressure levels were associated with higher atrial fibrillation risk (DBP: RR, 1.18; 95% CI, 1.03 to 1.35; P = 0.012. SBP: RR, 1.18; 95% CI, 1.01 to 1.38; P = 0.04). Specially, we also found the bidirectional causality between blood pressure and ischemic stroke. CONCLUSIONS: Our study provided a strong evidence that raised blood pressure in stroke patients increases the risk of atrial fibrillation and active acute blood pressure lowering can improve the outcome in ischemic stroke patients.

A novel path-specific effect statistic for identifying the differential specific paths in systems epidemiology.

BACKGROUND: Biological pathways play an important role in the occurrence, development and recovery of complex diseases, such as cancers, which are multifactorial complex diseases that are generally caused by mutation of multiple genes or dysregulation of pathways. RESULTS: We propose a path-specific effect statistic (PSE) to detect the differential specific paths under two conditions (e.g. case VS. control groups, exposure Vs. nonexposure groups). In observational studies, the path-specific effect can be obtained by separately calculating the average causal effect of each directed edge through adjusting for the parent nodes of nodes in the specific path and multiplying them under each condition. Theoretical proofs and a series of simulations are conducted to validate the path-specific effect statistic. Applications are also performed to evaluate its practical performances. A series of simulation studies show that the Type I error rates of PSE with Permutation tests are more stable at the nominal level 0.05 and can accurately detect the differential specific paths when comparing with other methods. Specifically, the power reveals an increasing trends with the enlargement of path-specific effects and its effect differences under two conditions. Besides, the power of PSE is robust to the variation of parent or child node of the nodes on specific paths. Application to real data of Glioblastoma Multiforme (GBM), we successfully identified 14 positive specific pathways in mTOR pathway contributing to survival time of patients with GBM. All codes for automatic searching specific paths linking two continuous variables and adjusting set as well as PSE statistic can be found in supplementary materials.  CONCLUSION: The proposed PSE statistic can accurately detect the differential specific pathways contributing to complex disease and thus potentially provides new insights and ways to unlock the black box of disease mechanisms.

Causal data fusion methods using summary-level statistics for a continuous outcome.

In many empirical studies, there exist rich individual studies to separately estimate causal effect of the treatment or exposure variable on the outcome variable, but incomplete confounders are adjusted in each study. Suppose we are interested in the causal effect of a treatment or exposure on an outcome variable, and we have available rich datasets that contain different confounders. How to integrate summary-level statistics from multiple individual datasets to improve causal inference has become a main challenge in data fusion. We propose a novel method in this article to identify the causal effect of a treatment or exposure on the continuous outcome. We show that the causal effect is identifiable and can be estimated by combining summary-level statistics from multiple datasets containing subsets of confounders and an external dataset only containing complete confounding information. Simulation studies indicate the unbiasedness of causal effect estimate by our method and we apply our method to a study about the effect of body mass index on fasting blood glucose.

A novel method for controlling unobserved confounding using double confounders.

BACKGROUND: Controlling unobserved confounding still remains a great challenge in observational studies, and a series of strict assumptions of the existing methods usually may be violated in practice. Therefore, it is urgent to put forward a novel method. METHODS: We are interested in the causal effect of an exposure on the outcome, which is always confounded by unobserved confounding. We show that, the causal effect of an exposure on a continuous or categorical outcome is nonparametrically identified through only two independent or correlated available confounders satisfying a non-linear condition on the exposure. Asymptotic theory and variance estimators are developed for each case. We also discuss an extension for more than two binary confounders. RESULTS: The simulations show better estimation performance by our approach in contrast to the traditional regression approach adjusting for observed confounders. A real application is separately applied to assess the effects of Body Mass Index (BMI) on Systolic Blood Pressure (SBP), Diastolic Blood Pressure (DBP), Fasting Blood Glucose (FBG), Triglyceride (TG), Total Cholesterol (TC), High Density Lipoprotein (HDL) and Low Density Lipoprotein (LDL) with individuals in Shandong Province, China. Our results suggest that SBP increased 1.60 (95% CI: 0.99-2.93) mmol/L with per 1- kg/m2 higher BMI and DBP increased 0.37 (95% CI: 0.03-0.76) mmol/L with per 1- kg/m2 higher BMI. Moreover, 1- kg/m2 increase in BMI was causally associated with a 1.61 (95% CI: 0.96-2.97) mmol/L increase in TC, a 1.66 (95% CI: 0.91-55.30) mmol/L increase in TG and a 2.01 (95% CI: 1.09-4.31) mmol/L increase in LDL. However, BMI was not causally associated with HDL with effect value - 0.20 (95% CI: - 1.71-1.44). And, the effect value of FBG per 1- kg/m2 higher BMI was 0.56 (95% CI: - 0.24-2.18). CONCLUSIONS: We propose a novel method to control unobserved confounders through double binary confounders satisfying a non-linear condition on the exposure which is easy to access.

[Evaluating the causal relationship between hip circumference and coronary heart disease: a Mendelian randomization study].

OBJECTIVE: To test the causal effect of hip circumference adjusted for body mass index(HCadjBMI) and coronary heart disease(CHD) using a Mendelian randomization analysis. METHODS: Based on genome-wide association study, the associations between the genetic instruments(IVs) and HCadjBMI were obtained from the GIANT consortium(n=211 114, European), the associations between IVs and CHD were derided from CARDIoGRAM consortium(n=86 995, European). The inverse-variance weighted method was used to estimate a pooled OR for the effect of a 1 cm higher HCadjBMI on CHD. Evidence of directional pleiotropy averaged across all variants was sought using MR-Egger regression. RESULTS: A total of 70 genetic variants that reached genome-wide significance and independent of each other were identified as IVs. A combined genetic variants expected to confer a lifetime exposure of per SD higher HCadjBMI was associated with a lower risk of CHD(OR=0. 831, 95%CI 0. 730-0. 946). MR-Egger regression intercept suggested that directional pleiotropy was unlikely to be biasing the result(intercept-0. 0012, P=0. 875). There was no specific single nucleotide polymorphism(SNP) detected by "leave one out" analysis. CONCLUSION: A genetic predisposition to higher HCadjBMI was associated with lower risk of CHD.

Why Do Testate Amoeba Optima Related to Water Table Depth Vary?

This study focusses on the ecology of testate amoeba species in peatlands of the southern taiga of Western Siberia. To estimate the influence of the trophic state of mires on species optima related to water table depth, a separate study of three calibration datasets including ombrotrophic, minerotrophic and the combined habitats was conducted. In the datasets obtained separately from ombrotrophic and minerotrophic mires, the water table depth was the main factor affecting testate amoeba assemblages. However, the trophic state (specifically pH and ash content) was more important factor in the combined dataset, including all of the studied mires. For 36 testate amoeba species, which were found in the ombrotrophic and minerotrophic mire habitats, their species optima, obtained separately in ombrotrophic and minerotrophic datasets, differed significantly from each other. Some of these species preferred minerotrophic conditions, while others preferred ombrotrophic ones. For all species, the trophic state of the mires affected the values of the species optima related to water table depth, as revealed in the form of a threshold effect. In extreme conditions, the species were more sensitive to the trophic status than to the water table depth, and their optimum related to water table depth was distorted. Variation of the optimum was observed in those species that inhabited both ombrotrophic and minerotrophic mires due to the fact that mires with a different trophic status were included in the training sets. The optima did not vary for species inhabiting only ombrotrophic or only minerotrophic mires.

Serum Metrnl Level is Correlated with Insulin Resistance, But Not with ß-Cell Function in Type 2 Diabetics.

BACKGROUND Metrnl is a novel identified adipomyokine which might have therapeutic potential for metabolic and inflammatory diseases, including type 2 diabetes mellitus. We aimed to explore the associations of circulating Metrnl level with ß-cell function and insulin resistance (IR) and further explore the possible correlation between Metrnl and another adipomyokine named irisin in patients diagnosed type 2 diabetes. MATERIAL AND METHODS Our study recruited 59 participants with type 2 diabetes and 30 normal glucose tolerance (NGT) participants. We used enzyme-linked immunosorbent assay (ELISA) to measure serum levels of Metrnl and irisin. The associations of Metrnl level with indexes of ß-cell function and IR and irisin level were analyzed by multiple linear regression analysis or spearman correlation analysis. RESULTS Compared with NGT participants, serum Metrnl level was elevated in participants with type 2 diabetes: 210.30 pg/mL (range 105.94-323.91 pg/mL) versus 132.02 pg/mL (range 104.93-195.92 pg/mL). Metrnl level did not show significant correlation with ß-cell function-related indicators, but positively correlated with HOMA2-IR and negatively correlated with HOMA2-%S after controlling multiple covariates in participants with type 2 diabetes. Metrnl level was also not associated with obesity-related indicators (body mass index, waist circumference, body fat percentage, and visceral adipose tissue area) in the type 2 diabetes group. In addition, the correlation between Metrnl and irisin level was also not present (r=-0.159, P=0.229) in type 2 diabetes group. CONCLUSIONS Serum Metrnl level was associated with IR, but not with ß-cell function in participants with diagnosed type 2 diabetes.

The alarming problems of confounding equivalence using logistic regression models in the perspective of causal diagrams.

BACKGROUND: Confounders can produce spurious associations between exposure and outcome in observational studies. For majority of epidemiologists, adjusting for confounders using logistic regression model is their habitual method, though it has some problems in accuracy and precision. It is, therefore, important to highlight the problems of logistic regression and search the alternative method. METHODS: Four causal diagram models were defined to summarize confounding equivalence. Both theoretical proofs and simulation studies were performed to verify whether conditioning on different confounding equivalence sets had the same bias-reducing potential and then to select the optimum adjusting strategy, in which logistic regression model and inverse probability weighting based marginal structural model (IPW-based-MSM) were compared. The "do-calculus" was used to calculate the true causal effect of exposure on outcome, then the bias and standard error were used to evaluate the performances of different strategies. RESULTS: Adjusting for different sets of confounding equivalence, as judged by identical Markov boundaries, produced different bias-reducing potential in the logistic regression model. For the sets satisfied G-admissibility, adjusting for the set including all the confounders reduced the equivalent bias to the one containing the parent nodes of the outcome, while the bias after adjusting for the parent nodes of exposure was not equivalent to them. In addition, all causal effect estimations through logistic regression were biased, although the estimation after adjusting for the parent nodes of exposure was nearest to the true causal effect. However, conditioning on different confounding equivalence sets had the same bias-reducing potential under IPW-based-MSM. Compared with logistic regression, the IPW-based-MSM could obtain unbiased causal effect estimation when the adjusted confounders satisfied G-admissibility and the optimal strategy was to adjust for the parent nodes of outcome, which obtained the highest precision. CONCLUSIONS: All adjustment strategies through logistic regression were biased for causal effect estimation, while IPW-based-MSM could always obtain unbiased estimation when the adjusted set satisfied G-admissibility. Thus, IPW-based-MSM was recommended to adjust for confounders set.

A novel Markov Blanket-based repeated-fishing strategy for capturing phenotype-related biomarkers in big omics data.

BACKGROUND: We propose a novel Markov Blanket-based repeated-fishing strategy (MBRFS) in attempt to increase the power of existing Markov Blanket method (DASSO-MB) and maintain its advantages in omic data analysis. RESULTS: Both simulation and real data analysis were conducted to assess its performances by comparing with other methods including χ(2) test with Bonferroni and B-H adjustment, least absolute shrinkage and selection operator (LASSO) and DASSO-MB. A serious of simulation studies showed that the true discovery rate (TDR) of proposed MBRFS was always close to zero under null hypothesis (odds ratio = 1 for each SNPs) with excellent stability in all three scenarios of independent phenotype-related SNPs without linkage disequilibrium (LD) around them, correlated phenotype-related SNPs without LD around them, and phenotype-related SNPs with strong LD around them. As expected, under different odds ratio and minor allel frequency (MAFs), MBRFS always had the best performances in capturing the true phenotype-related biomarkers with higher matthews correlation coefficience (MCC) for all three scenarios above. More importantly, since proposed MBRFS using the repeated fishing strategy, it still captures more phenotype-related SNPs with minor effects when non-significant phenotype-related SNPs emerged under χ(2) test after Bonferroni multiple correction. The various real omics data analysis, including GWAS data, DNA methylation data, gene expression data and metabolites data, indicated that the proposed MBRFS always detected relatively reasonable biomarkers. CONCLUSIONS: Our proposed MBRFS can exactly capture the true phenotype-related biomarkers with the reduction of false negative rate when the phenotype-related biomarkers are independent or correlated, as well as the circumstance that phenotype-related biomarkers are associated with non-phenotype-related ones.

A powerful score-based test statistic for detecting gene-gene co-association.

BACKGROUND: The genetic variants identified by Genome-wide association study (GWAS) can only account for a small proportion of the total heritability for complex disease. The existence of gene-gene joint effects which contains the main effects and their co-association is one of the possible explanations for the "missing heritability" problems. Gene-gene co-association refers to the extent to which the joint effects of two genes differ from the main effects, not only due to the traditional interaction under nearly independent condition but the correlation between genes. Generally, genes tend to work collaboratively within specific pathway or network contributing to the disease and the specific disease-associated locus will often be highly correlated (e.g. single nucleotide polymorphisms (SNPs) in linkage disequilibrium). Therefore, we proposed a novel score-based statistic (SBS) as a gene-based method for detecting gene-gene co-association. RESULTS: Various simulations illustrate that, under different sample sizes, marginal effects of causal SNPs and co-association levels, the proposed SBS has the better performance than other existed methods including single SNP-based and principle component analysis (PCA)-based logistic regression model, the statistics based on canonical correlations (CCU), kernel canonical correlation analysis (KCCU), partial least squares path modeling (PLSPM) and delta-square (δ (2)) statistic. The real data analysis of rheumatoid arthritis (RA) further confirmed its advantages in practice. CONCLUSIONS: SBS is a powerful and efficient gene-based method for detecting gene-gene co-association.