The challenge for genetic epidemiologists: how to analyze large numbers of SNPs in relation to complex diseases

Enhancing mass spectrometry imaging accessibility using convolutional autoencoders for deriving hypoxia-associated peptides from tumors

Mass spectrometry imaging (MSI) allows to study cancer's intratumoral heterogeneity through spatially-resolved peptides, metabolites and lipids. Yet, in biomedical research MSI is rarely used for biomarker discovery. Besides its high dimensionality and multicollinearity, mass spectrometry (MS) technologies typically output mass-to-charge ratio values but not the biochemical compounds of interest. Our framework makes particularly low-abundant signals in MSI more accessible. We utilized convolutional autoencoders to aggregate features associated with tumor hypoxia, a parameter with significant spatial heterogeneity, in cancer xenograft models. We highlight that MSI captures these low-abundant signals and that autoencoders can preserve them in their latent space. The relevance of individual hyperparameters is demonstrated through ablation experiments, and the contribution from original features to latent features is unraveled. Complementing MSI with tandem MS from the same tumor model, multiple hypoxia-associated peptide candidates were derived. Compared to random forests alone, our autoencoder approach yielded more biologically relevant insights for biomarker discovery.

Genetic risk assessment based on association and prediction studies

The genetic basis of phenotypic emergence provides valuable information for assessing individual risk. While association studies have been pivotal in identifying genetic risk factors within a population, complementing it with insights derived from predictions studies that assess individual-level risk offers a more comprehensive approach to understanding phenotypic expression. In this study, we established personalized risk assessment models using single-nucleotide polymorphism (SNP) data from 200 Korean patients, of which 100 experienced hepatitis B surface antigen (HBsAg) seroclearance and 100 patients demonstrated high levels of HBsAg. The risk assessment models determined the predictive power of the following: (1) genome-wide association study (GWAS)-identified candidate biomarkers considered significant in a reference study and (2) machine learning (ML)-identified candidate biomarkers with the highest feature importance scores obtained by using random forest (RF). While utilizing all features yielded 64% model accuracy, using relevant biomarkers achieved higher model accuracies: 82% for 52 GWAS-identified candidate biomarkers, 71% for three GWAS-identified biomarkers, and 80% for 150 ML-identified candidate biomarkers. Findings highlight that the joint contributions of relevant biomarkers significantly influence phenotypic emergence. On the other hand, combining ML-identified candidate biomarkers into the pool of GWAS-identified candidate biomarkers resulted in the improved predictive accuracy of 90%, demonstrating the capability of ML as an auxiliary analysis to GWAS. Furthermore, some of the ML-identified candidate biomarkers were found to be linked with hepatocellular carcinoma (HCC), reinforcing previous claims that HCC can still occur despite the absence of HBsAg.

Synergism interaction between genetic polymorphisms in drug metabolizing enzymes and NSAIDs on upper gastrointestinal haemorrhage: a multicenter case-control study

BACKGROUND

Interindividual genetic variations contribute to differences in patients' response to drugs as well as to the development of certain disorders. Patients who use non-steroidal anti-inflammatory drugs (NSAIDs) may develop serious gastrointestinal disorders, mainly upper gastrointestinal haemorrhage (UGIH). Studies about the interaction between NSAIDs and genetic variations on the risk of UGIH are scarce. Therefore, we investigated the effect of 16 single nucleotide polymorphisms (SNPs) involved in drug metabolism on the risk of NSAIDs-induced UGIH.

MATERIALS AND METHODS

We conducted a multicenter case-control study of 326 cases and 748 controls. Participants were sub-grouped into four categories according to NSAID exposure and genetic profile. We estimated odds ratios (ORs) and their 95% confidence intervals (CI) using generalized linear mixed models for dependent binomial variables and then calculated the measures of interaction, synergism index (S), and relative excess risk due to interaction (RERI). We undertook stratified analyses by the type of NSAID (aspirin, non-aspirin).

RESULTS

We observed an excess risk of UGIH due to an interaction between any NSAID, non-aspirin NSAIDs or aspirin and carrying certain SNPs. The greatest excess risk was observed for carriers of: rs2180314:C>G [any NSAID: S = 3.30 (95%CI: 1.24-8.80), RERI = 4.39 (95%CI: 0.70-8.07); non-aspirin NSAIDs: S = 3.42 (95%CI: 1.12-10.47), RERI = 3.97 (95%CI: 0.44-7.50)], and rs4809957:A>G [any NSAID: S = 2.11 (95%CI: 0.90-4.97), RERI = 3.46 (95%CI: -0.40-7.31)]. Aspirin use by carriers of rs6664:C>T is also associated with increased risk of UGIH [OR_{aspirin(+),wild-type}: 2.22 (95%CI: 0.69-7.17) vs. OR_{aspirin(+),genetic-variation}: 7.72 (95%CI: 2.75-21.68)], yet larger sample size is needed to confirm this observation.

CONCLUSIONS

The joint effect of the SNPs s2180314:C>G and rs4809957:A>G and NSAIDs are more than three times higher than the sum of their individual effects. Personalized prescriptions based on genotyping would permit a better weighing of risks and benefits from NSAID consumption.KEY MESSAGESMulticenter case-control study of the effect of genetic variations involved in drug metabolism on upper gastrointestinal haemorrhage (UGIH) induced by NSAIDs (aspirin and non-aspirin).There is a statistically significant additive synergism interaction between certain genetic polymorphisms and NSAIDs on UGIH: rs2180314:C>G and rs4809957:A>G. The joint effect of each of these single nucleotide polymorphisms and NSAIDs on UGIH is more than three times higher than the sum of their individual effects.Genetic profiling and personalized prescriptions would be useful in managing the risks and benefits associated with NSAIDs.

Genetic heterogeneity: Challenges, impacts, and methods through an associative lens

Genetic heterogeneity describes the occurrence of the same or similar phenotypes through different genetic mechanisms in different individuals. Robustly characterizing and accounting for genetic heterogeneity is crucial to pursuing the goals of precision medicine, for discovering novel disease biomarkers, and for identifying targets for treatments. Failure to account for genetic heterogeneity may lead to missed associations and incorrect inferences. Thus, it is critical to review the impact of genetic heterogeneity on the design and analysis of population level genetic studies, aspects that are often overlooked in the literature. In this review, we first contextualize our approach to genetic heterogeneity by proposing a high-level categorization of heterogeneity into "feature," "outcome," and "associative" heterogeneity, drawing on perspectives from epidemiology and machine learning to illustrate distinctions between them. We highlight the unique nature of genetic heterogeneity as a heterogeneous pattern of association that warrants specific methodological considerations. We then focus on the challenges that preclude effective detection and characterization of genetic heterogeneity across a variety of epidemiological contexts. Finally, we discuss systems heterogeneity as an integrated approach to using genetic and other high-dimensional multi-omic data in complex disease research.

Identification of miRNA-disease associations via deep forest ensemble learning based on autoencoder

Increasing evidences show that the occurrence of human complex diseases is closely related to microRNA (miRNA) variation and imbalance. For this reason, predicting disease-related miRNAs is essential for the diagnosis and treatment of complex human diseases. Although some current computational methods can effectively predict potential disease-related miRNAs, the accuracy of prediction should be further improved. In our study, a new computational method via deep forest ensemble learning based on autoencoder (DFELMDA) is proposed to predict miRNA-disease associations. Specifically, a new feature representation strategy is proposed to obtain different types of feature representations (from miRNA and disease) for each miRNA-disease association. Then, two types of low-dimensional feature representations are extracted by two deep autoencoders for predicting miRNA-disease associations. Finally, two prediction scores of the miRNA-disease associations are obtained by the deep random forest and combined to determine the final results. DFELMDA is compared with several classical methods on the The Human microRNA Disease Database (HMDD) dataset. Results reveal that the performance of this method is superior. The area under receiver operating characteristic curve (AUC) values obtained by DFELMDA through 5-fold and 10-fold cross-validation are 0.9552 and 0.9560, respectively. In addition, case studies on colon, breast and lung tumors of different disease types further demonstrate the excellent ability of DFELMDA to predict disease-associated miRNA-disease. Performance analysis shows that DFELMDA can be used as an effective computational tool for predicting miRNA-disease associations.

A Machine Learning Tool to Predict the Antibacterial Capacity of Nanoparticles

The emergence and rapid spread of multidrug-resistant bacteria strains are a public health concern. This emergence is caused by the overuse and misuse of antibiotics leading to the evolution of antibiotic-resistant strains. Nanoparticles (NPs) are objects with all three external dimensions in the nanoscale that varies from 1 to 100 nm. Research on NPs with enhanced antimicrobial activity as alternatives to antibiotics has grown due to the increased incidence of nosocomial and community acquired infections caused by pathogens. Machine learning (ML) tools have been used in the field of nanoinformatics with promising results. As a consequence of evident achievements on a wide range of predictive tasks, ML techniques are attracting significant interest across a variety of stakeholders. In this article, we present an ML tool that successfully predicts the antibacterial capacity of NPs while the model's validation demonstrates encouraging results (R2 = 0.78). The data were compiled after a literature review of 60 articles and consist of key physico-chemical (p-chem) properties and experimental conditions (exposure variables and bacterial clustering) from in vitro studies. Following data homogenization and pre-processing, we trained various regression algorithms and we validated them using diverse performance metrics. Finally, an important attribute evaluation, which ranks the attributes that are most important in predicting the outcome, was performed. The attribute importance revealed that NP core size, the exposure dose, and the species of bacterium are key variables in predicting the antibacterial effect of NPs. This tool assists various stakeholders and scientists in predicting the antibacterial effects of NPs based on their p-chem properties and diverse exposure settings. This concept also aids the safe-by-design paradigm by incorporating functionality tools.

Machine learning approaches for the prediction of bone mineral density by using genomic and phenotypic data of 5130 older men

The study aimed to utilize machine learning (ML) approaches and genomic data to develop a prediction model for bone mineral density (BMD) and identify the best modeling approach for BMD prediction. The genomic and phenotypic data of Osteoporotic Fractures in Men Study (n = 5130) was analyzed. Genetic risk score (GRS) was calculated from 1103 associated SNPs for each participant after a comprehensive genotype imputation. Data were normalized and divided into a training set (80%) and a validation set (20%) for analysis. Random forest, gradient boosting, neural network, and linear regression were used to develop BMD prediction models separately. Ten-fold cross-validation was used for hyper-parameters optimization. Mean square error and mean absolute error were used to assess model performance. When using GRS and phenotypic covariates as the predictors, all ML models' performance and linear regression in BMD prediction were similar. However, when replacing GRS with the 1103 individual SNPs in the model, ML models performed significantly better than linear regression (with lasso regularization), and the gradient boosting model performed the best. Our study suggested that ML models, especially gradient boosting, can improve BMD prediction in genomic data.

Personizing the prediction of future susceptibility to a specific disease

A traceable biomarker is a member of a disease's molecular pathway. A disease may be associated with several molecular pathways. Each different combination of these molecular pathways, to which detected traceable biomarkers belong, may serve as an indicative of the elicitation of the disease at a different time frame in the future. Based on this notion, we introduce a novel methodology for personalizing an individual's degree of future susceptibility to a specific disease. We implemented the methodology in a working system called Susceptibility Degree to a Disease Predictor (SDDP). For a specific disease d, let S be the set of molecular pathways, to which traceable biomarkers detected from most patients of d belong. For the same disease d, let S' be the set of molecular pathways, to which traceable biomarkers detected from a certain individual belong. SDDP is able to infer the subset S'' ⊆{S-S'} of undetected molecular pathways for the individual. Thus, SDDP can infer undetected molecular pathways of a disease for an individual based on few molecular pathways detected from the individual. SDDP can also help in inferring the combination of molecular pathways in the set {S'+S''}, whose traceable biomarkers collectively is an indicative of the disease. SDDP is composed of the following four components: information extractor, interrelationship between molecular pathways modeler, logic inferencer, and risk indicator. The information extractor takes advantage of the exponential increase of biomedical literature to automatically extract the common traceable biomarkers for a specific disease. The interrelationship between molecular pathways modeler models the hierarchical interrelationships between the molecular pathways of the traceable biomarkers. The logic inferencer transforms the hierarchical interrelationships between the molecular pathways into rule-based specifications. It employs the specification rules and the inference rules for predicate logic to infer as many as possible undetected molecular pathways of a disease for an individual. The risk indicator outputs a risk indicator value that reflects the individual's degree of future susceptibility to the disease. We evaluated SDDP by comparing it experimentally with other methods. Results revealed marked improvement.

Identification of early liver toxicity gene biomarkers using comparative supervised machine learning

Screening agrochemicals and pharmaceuticals for potential liver toxicity is required for regulatory approval and is an expensive and time-consuming process. The identification and utilization of early exposure gene signatures and robust predictive models in regulatory toxicity testing has the potential to reduce time and costs substantially. In this study, comparative supervised machine learning approaches were applied to the rat liver TG-GATEs dataset to develop feature selection and predictive testing. We identified ten gene biomarkers using three different feature selection methods that predicted liver necrosis with high specificity and selectivity in an independent validation dataset from the Microarray Quality Control (MAQC)-II study. Nine of the ten genes that were selected with the supervised methods are involved in metabolism and detoxification (Car3, Crat, Cyp39a1, Dcd, Lbp, Scly, Slc23a1, and Tkfc) and transcriptional regulation (Ablim3). Several of these genes are also implicated in liver carcinogenesis, including Crat, Car3 and Slc23a1. Our biomarker gene signature provides high statistical accuracy and a manageable number of genes to study as indicators to potentially accelerate toxicity testing based on their ability to induce liver necrosis and, eventually, liver cancer.

Personalized Psychotherapy for Outpatients with Major Depression and Anxiety Disorders: Transdiagnostic Versus Diagnosis-Specific Group Cognitive Behavioural Therapy

Background

Only about half of all patients with anxiety disorders or major depression respond to cognitive behaviour therapy (CBT), even though this is an evidence-based treatment. Personalized treatment offers an approach to increase the number of patients who respond to therapy. The aim of this study was to examine predictors and moderators of (differential) treatment outcomes in transdiagnostic versus diagnosis-specific group CBT.

Methods

A sample of 291 patients from three different mental health clinics in Denmark was randomized to either transdiagnostic or diagnosis-specific group CBT. The study outcome was the regression slope of the individual patient's repeated scores on the WHO-5 Well-being Index. Pre-treatment variables were identified as moderators or predictors through a two-step variable selection approach.

Results

While the two-step approach failed to identify any moderators, four predictors were found: level of positive affect, duration of disorder, the detachment personality trait, and the coping strategy of cognitive reappraisal. A prognostic index was constructed, but did not seem to be robust across treatment sites.

Conclusions

Our findings give insufficient evidence to support a recommendation of either transdiagnostic or diagnosis-specific CBT for a given patient or to predict the response to the applied group therapies.

A random forest based computational model for predicting novel lncRNA-disease associations

BACKGROUND

Accumulated evidence shows that the abnormal regulation of long non-coding RNA (lncRNA) is associated with various human diseases. Accurately identifying disease-associated lncRNAs is helpful to study the mechanism of lncRNAs in diseases and explore new therapies of diseases. Many lncRNA-disease association (LDA) prediction models have been implemented by integrating multiple kinds of data resources. However, most of the existing models ignore the interference of noisy and redundancy information among these data resources.

RESULTS

To improve the ability of LDA prediction models, we implemented a random forest and feature selection based LDA prediction model (RFLDA in short). First, the RFLDA integrates the experiment-supported miRNA-disease associations (MDAs) and LDAs, the disease semantic similarity (DSS), the lncRNA functional similarity (LFS) and the lncRNA-miRNA interactions (LMI) as input features. Then, the RFLDA chooses the most useful features to train prediction model by feature selection based on the random forest variable importance score that takes into account not only the effect of individual feature on prediction results but also the joint effects of multiple features on prediction results. Finally, a random forest regression model is trained to score potential lncRNA-disease associations. In terms of the area under the receiver operating characteristic curve (AUC) of 0.976 and the area under the precision-recall curve (AUPR) of 0.779 under 5-fold cross-validation, the performance of the RFLDA is better than several state-of-the-art LDA prediction models. Moreover, case studies on three cancers demonstrate that 43 of the 45 lncRNAs predicted by the RFLDA are validated by experimental data, and the other two predicted lncRNAs are supported by other LDA prediction models.

CONCLUSIONS

Cross-validation and case studies indicate that the RFLDA has excellent ability to identify potential disease-associated lncRNAs.

Precision immunoprofiling to reveal diagnostic signatures for latent tuberculosis infection and reactivation risk stratification

Latent tuberculosis infection (LTBI) is estimated in nearly one quarter of the world's population, and of those immunocompetent and infected ~10% will proceed to active tuberculosis (TB). Current diagnostics cannot definitively identify LTBI and provide no insight into reactivation risk, thereby defining an unmet diagnostic challenge of incredible global significance. We introduce a new machine-learning-driven approach to LTBI diagnostics that leverages a high throughput, multiplexed cytokine detection technology and powerful bioinformatics to reveal multi-marker signatures for LTBI diagnosis and risk stratification. This approach is enabled through an individualized normalization procedure that allows disease-relevant biomarker signatures to be revealed despite heterogeneity in basal immune response. Specifically, cytokines secreted from antigen-challenged peripheral blood mononuclear cells were detected using silicon photonic sensor arrays and multidimensional data correlation of individually-normalized immune responses revealed signatures important for LTBI status. These results demonstrate a powerful combination of multiplexed biomarker detection technologies, precision immune normalization, and feature selection algorithms that revealed positively correlated multi-biomarker signatures for LTBI status and reactivation risk stratification from a relatively simple blood-based assay.

Altered Hippocampal Epigenetic Regulation Underlying Reduced Cognitive Development in Response to Early Life Environmental Insults

The hippocampus is involved in learning and memory and undergoes significant growth and maturation during the neonatal period. Environmental insults during this developmental timeframe can have lasting effects on brain structure and function. This study assessed hippocampal DNA methylation and gene transcription from two independent studies reporting reduced cognitive development stemming from early life environmental insults (iron deficiency and porcine reproductive and respiratory syndrome virus (PRRSv) infection) using porcine biomedical models. In total, 420 differentially expressed genes (DEGs) were identified between the reduced cognition and control groups, including genes involved in neurodevelopment and function. Gene ontology (GO) terms enriched for DEGs were associated with immune responses, angiogenesis, and cellular development. In addition, 116 differentially methylated regions (DMRs) were identified, which overlapped 125 genes. While no GO terms were enriched for genes overlapping DMRs, many of these genes are known to be involved in neurodevelopment and function, angiogenesis, and immunity. The observed altered methylation and expression of genes involved in neurological function suggest reduced cognition in response to early life environmental insults is due to altered cholinergic signaling and calcium regulation. Finally, two DMRs overlapped with two DEGs, VWF and LRRC32, which are associated with blood brain barrier permeability and regulatory T-cell activation, respectively. These results support the role of altered hippocampal DNA methylation and gene expression in early life environmentally-induced reductions in cognitive development across independent studies.

Machine learning approach to single nucleotide polymorphism-based asthma prediction

Machine learning (ML) is poised as a transformational approach uniquely positioned to discover the hidden biological interactions for better prediction and diagnosis of complex diseases. In this work, we integrated ML-based models for feature selection and classification to quantify the risk of individual susceptibility to asthma using single nucleotide polymorphism (SNP). Random forest (RF) and recursive feature elimination (RFE) algorithm were implemented to identify the SNPs with high implication to asthma. K-nearest neighbor (kNN) and support vector machine (SVM) algorithms were trained to classify the identified SNPs whether associated with non-asthmatic or asthmatic samples. Feature selection step showed that RF outperformed RFE and the feature importance score derived from RF was consistently high for a subset of SNPs, indicating the robustness of RF in selecting relevant features associated with asthma. Model comparison showed that the integration of RF-SVM obtained the highest model performance with an accuracy, precision, and sensitivity of 62.5%, 65.3%, and 69%, respectively, when compared to the baseline, RF-kNN, and an external MeanDiff-kNN models. Furthermore, results show that the occurrence of asthma can be predicted with an Area under the Curve (AUC) of 0.62 and 0.64 for RF-SVM and RF-kNN models, respectively. This study demonstrates the integration of ML models to augment traditional methods in predicting genetic predisposition to multifactorial diseases such as asthma.

Validating a continental-scale groundwater diffuse pollution model using regional datasets

In this study, we assess the validity of an African-scale groundwater pollution model for nitrates. In a previous study, we identified a statistical continental-scale groundwater pollution model for nitrate. The model was identified using a pan-African meta-analysis of available nitrate groundwater pollution studies. The model was implemented in both Random Forest (RF) and multiple regression formats. For both approaches, we collected as predictors a comprehensive GIS database of 13 spatial attributes, related to land use, soil type, hydrogeology, topography, climatology, region typology, nitrogen fertiliser application rate, and population density. In this paper, we validate the continental-scale model of groundwater contamination by using a nitrate measurement dataset from three African countries. We discuss the issue of data availability, and quality and scale issues, as challenges in validation. Notwithstanding that the modelling procedure exhibited very good success using a continental-scale dataset (e.g. R² = 0.97 in the RF format using a cross-validation approach), the continental-scale model could not be used without recalibration to predict nitrate pollution at the country scale using regional data. In addition, when recalibrating the model using country-scale datasets, the order of model exploratory factors changes. This suggests that the structure and the parameters of a statistical spatially distributed groundwater degradation model for the African continent are strongly scale dependent.

SNP by SNP by environment interaction network of alcoholism

Background

Alcoholism has a strong genetic component. Twin studies have demonstrated the heritability of a large proportion of phenotypic variance of alcoholism ranging from 50–80%. The search for genetic variants associated with this complex behavior has epitomized sequence-based studies for nearly a decade. The limited success of genome-wide association studies (GWAS), possibly precipitated by the polygenic nature of complex traits and behaviors, however, has demonstrated the need for novel, multivariate models capable of quantitatively capturing interactions between a host of genetic variants and their association with non-genetic factors. In this regard, capturing the network of SNP by SNP or SNP by environment interactions has recently gained much interest.

Results

Here, we assessed 3,776 individuals to construct a network capable of detecting and quantifying the interactions within and between plausible genetic and environmental factors of alcoholism. In this regard, we propose the use of first-order dependence tree of maximum weight as a potential statistical learning technique to delineate the pattern of dependencies underpinning such a complex trait. Using a predictive based analysis, we further rank the genes, demographic factors, biological pathways, and the interactions represented by our SNP \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ \times $$\end{document}×SNP\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ \times $$\end{document}×E network. The proposed framework is quite general and can be potentially applied to the study of other complex traits. 

Electronic supplementary material

The online version of this article (doi:10.1186/s12918-017-0403-7) contains supplementary material, which is available to authorized users.

Allelic Interactions among Pto-MIR475b and Its Four Target Genes Potentially Affect Growth and Wood Properties in Populus

MicroRNAs (miRNAs) play crucial roles in plant growth and development, but few studies have illuminated the allelic interactions among miRNAs and their targets in perennial plants. Here, we combined analysis of expression patterns and single-nucleotide polymorphism (SNP)-based association studies to explore the interactions between Pto-MIR475b and its four target genes (Pto-PPR1, Pto-PPR2, Pto-PPR3, and Pto-PPR4) in 435 unrelated individuals of Populus tomentosa. Expression patterns showed a significant negative correlation (r = -0.447 to -0.411, P < 0.01) between Pto-MIR475b and its four targets in eight tissues of P. tomentosa, suggesting that Pto-miR475b may negatively regulate the four targets. Single SNP-based association studies identified 93 significant associations (P < 0.01, Q < 0.1) representing associations of 80 unique SNPs in Pto-MIR475b and its four targets with nine traits, revealing their potential roles in tree growth and wood formation. Moreover, one common SNP in the precursor region significantly altered the secondary structure of the pre-Pto-miR475b and changed the expression level of Pto-MIR475b. Analysis of epistatic interactions identified 115 significant SNP-SNP associations (P < 0.01) representing 45 unique SNPs from Pto-MIR475b and its four targets for 10 traits, revealing that genetic interactions between Pto-MIR475b and its targets influence quantitative traits of perennial plants. Our study provided a feasible strategy to study population genetics in forest trees and enhanced our understanding of miRNAs by dissecting the allelic interactions between this miRNA and its targets in P. tomentosa.

SNP-SNP Interaction Analysis on Soybean Oil Content under Multi-Environments

Soybean oil content is one of main quality traits. In this study, we used the multifactor dimensionality reduction (MDR) method and a soybean high-density genetic map including 5,308 markers to identify stable single nucleotide polymorphism (SNP)—SNP interactions controlling oil content in soybean across 23 environments. In total, 36,442,756 SNP-SNP interaction pairs were detected, 1865 of all interaction pairs associated with soybean oil content were identified under multiple environments by the Bonferroni correction with p <3.55×10⁻¹¹. Two and 1863 SNP-SNP interaction pairs detected stable across 12 and 11 environments, respectively, which account around 50% of total environments. Epistasis values and contribution rates of stable interaction (the SNP interaction pairs were detected in more than 2 environments) pairs were detected by the two way ANOVA test, the available interaction pairs were ranged 0.01 to 0.89 and from 0.01 to 0.85, respectively. Some of one side of the interaction pairs were identified with previously research as a major QTL without epistasis effects. The results of this study provide insights into the genetic architecture of soybean oil content and can serve as a basis for marker-assisted selection breeding.

Evaluation of potential novel variations and their interactions related to bipolar disorders: analysis of genome-wide association study data

BACKGROUND

Multifactor dimensionality reduction (MDR) is a nonparametric approach that can be used to detect relevant interactions between single-nucleotide polymorphisms (SNPs). The aim of this study was to build the best genomic model based on SNP associations and to identify candidate polymorphisms that are the underlying molecular basis of the bipolar disorders.

METHODS

This study was performed on Whole-Genome Association Study of Bipolar Disorder (dbGaP [database of Genotypes and Phenotypes] study accession number: phs000017.v3.p1) data. After preprocessing of the genotyping data, three classification-based data mining methods (ie, random forest, naïve Bayes, and k-nearest neighbor) were performed. Additionally, as a nonparametric, model-free approach, the MDR method was used to evaluate the SNP profiles. The validity of these methods was evaluated using true classification rate, recall (sensitivity), precision (positive predictive value), and F-measure.

RESULTS

Random forests, naïve Bayes, and k-nearest neighbors identified 16, 13, and ten candidate SNPs, respectively. Surprisingly, the top six SNPs were reported by all three methods. Random forests and k-nearest neighbors were more successful than naïve Bayes, with recall values >0.95. On the other hand, MDR generated a model with comparable predictive performance based on five SNPs. Although different SNP profiles were identified in MDR compared to the classification-based models, all models mapped SNPs to the DOCK10 gene.

CONCLUSION

Three classification-based data mining approaches, random forests, naïve Bayes, and k-nearest neighbors, have prioritized similar SNP profiles as predictors of bipolar disorders, in contrast to MDR, which has found different SNPs through analysis of two-way and three-way interactions. The reduced number of associated SNPs discovered by MDR, without loss in the classification performance, would facilitate validation studies and decision support models, and would reduce the cost to develop predictive and diagnostic tests. Nevertheless, we need to emphasize that translation of genomic models to the clinical setting requires models with higher classification performance.

Comparing high-dimensional confounder control methods for rapid cohort studies from electronic health records

Aims:

Electronic health records (EHR), containing rich clinical histories of large patient populations, can provide evidence for clinical decisions when evidence from trials and literature is absent. To enable such observational studies from EHR in real time, particularly in emergencies, rapid confounder control methods that can handle numerous variables and adjust for biases are imperative. This study compares the performance of 18 automatic confounder control methods.

Methods:

Methods include propensity scores, direct adjustment by machine learning, similarity matching and resampling in two simulated and one real-world EHR datasets.

Results & conclusions:

Direct adjustment by lasso regression and ensemble models involving multiple resamples have performance comparable to expert-based propensity scores and thus, may help provide real-time EHR-based evidence for timely clinical decisions.

Applicability of gene expression and systems biology to develop pharmacogenetic predictors; antipsychotic-induced extrapyramidal symptoms as an example

Pharmacogenetics has been driven by a candidate gene approach. The disadvantage of this approach is that is limited by our current understanding of the mechanisms by which drugs act. Gene expression could help to elucidate the molecular signatures of antipsychotic treatments searching for dysregulated molecular pathways and the relationships between gene products, especially protein-protein interactions. To embrace the complexity of drug response, machine learning methods could help to identify gene-gene interactions and develop pharmacogenetic predictors of drug response. The present review summarizes the applicability of the topics presented here (gene expression, network analysis and gene-gene interactions) in pharmacogenetics. In order to achieve this, we present an example of identifying genetic predictors of extrapyramidal symptoms induced by antipsychotic.

Research on single nucleotide polymorphisms interaction detection from network perspective

Single Nucleotide Polymorphisms (SNPs) found in Genome-Wide Association Study (GWAS) mainly influence the susceptibility of complex diseases, but they still could not comprehensively explain the relationships between mutations and diseases. Interactions between SNPs are considered so important for deeply understanding of those relationships that several strategies have been proposed to explore such interactions. However, part of those methods perform poorly when marginal effects of disease loci are weak or absent, others may lack of considering high-order SNPs interactions, few methods have achieved the requirements in both performance and accuracy. Considering the above reasons, not only low-order, but also high-order SNP interactions as well as main-effect SNPs, should be taken into account in detection methods under an acceptable computational complexity. In this paper, a new pairwise (or low-order) interaction detection method IG (Interaction Gain) is introduced, in which disease models are not required and parallel computing is utilized. Furthermore, high-order SNP interactions were proposed to be detected by finding closely connected function modules of the network constructed from IG detection results. Tested by a wide range of simulated datasets and four WTCCC real datasets, the proposed methods accurately detected both low-order and high-order SNP interactions as well as disease-associated main-effect SNPS and it surpasses all competitors in performances. The research will advance complex diseases research by providing more reliable SNP interactions.

Multivariate dimensionality reduction approaches to identify gene-gene and gene-environment interactions underlying multiple complex traits

The elusive but ubiquitous multifactor interactions represent a stumbling block that urgently needs to be removed in searching for determinants involved in human complex diseases. The dimensionality reduction approaches are a promising tool for this task. Many complex diseases exhibit composite syndromes required to be measured in a cluster of clinical traits with varying correlations and/or are inherently longitudinal in nature (changing over time and measured dynamically at multiple time points). A multivariate approach for detecting interactions is thus greatly needed on the purposes of handling a multifaceted phenotype and longitudinal data, as well as improving statistical power for multiple significance testing via a two-stage testing procedure that involves a multivariate analysis for grouped phenotypes followed by univariate analysis for the phenotypes in the significant group(s). In this article, we propose a multivariate extension of generalized multifactor dimensionality reduction (GMDR) based on multivariate generalized linear, multivariate quasi-likelihood and generalized estimating equations models. Simulations and real data analysis for the cohort from the Study of Addiction: Genetics and Environment are performed to investigate the properties and performance of the proposed method, as compared with the univariate method. The results suggest that the proposed multivariate GMDR substantially boosts statistical power.

Informative Bayesian Model Selection: a method for identifying interactions in genome-wide data

In high-dimensional genome-wide (GWA) data, a key challenge is to detect genomic variants that interact in a nonlinear fashion in their association with disease. Identifying such genomic interactions is important for elucidating the inheritance of complex phenotypes and diseases. In this paper, we introduce a new computational method called Informative Bayesian Model Selection (IBMS) that leverages correlation among variants in GWA data due to the linkage disequilibrium to identify interactions accurately in a computationally efficient manner. IBMS combines several statistical methods including canonical correlation analysis, logistic regression analysis, and a Bayesians statistical measure of evaluating interactions. Compared to BOOST and BEAM that are two widely used methods for detecting genomic interactions, IBMS had significantly higher power when evaluated on synthetic data. Furthermore, when applied to Alzheimer's disease GWA data, IBMS identified previously reported interactions. IBMS is a useful method for identifying variants in GWA data, and software that implements IBMS is freely available online from http://lbb.ut.ac.ir/Download/LBBsoft/IBMS.

Big data analysis using modern statistical and machine learning methods in medicine

In this article we introduce modern statistical machine learning and bioinformatics approaches that have been used in learning statistical relationships from big data in medicine and behavioral science that typically include clinical, genomic (and proteomic) and environmental variables. Every year, data collected from biomedical and behavioral science is getting larger and more complicated. Thus, in medicine, we also need to be aware of this trend and understand the statistical tools that are available to analyze these datasets. Many statistical analyses that are aimed to analyze such big datasets have been introduced recently. However, given many different types of clinical, genomic, and environmental data, it is rather uncommon to see statistical methods that combine knowledge resulting from those different data types. To this extent, we will introduce big data in terms of clinical data, single nucleotide polymorphism and gene expression studies and their interactions with environment. In this article, we will introduce the concept of well-known regression analyses such as linear and logistic regressions that has been widely used in clinical data analyses and modern statistical models such as Bayesian networks that has been introduced to analyze more complicated data. Also we will discuss how to represent the interaction among clinical, genomic, and environmental data in using modern statistical models. We conclude this article with a promising modern statistical method called Bayesian networks that is suitable in analyzing big data sets that consists with different type of large data from clinical, genomic, and environmental data. Such statistical model form big data will provide us with more comprehensive understanding of human physiology and disease.

Exploiting SNP correlations within random forest for genome-wide association studies

The primary goal of genome-wide association studies (GWAS) is to discover variants that could lead, in isolation or in combination, to a particular trait or disease. Standard approaches to GWAS, however, are usually based on univariate hypothesis tests and therefore can account neither for correlations due to linkage disequilibrium nor for combinations of several markers. To discover and leverage such potential multivariate interactions, we propose in this work an extension of the Random Forest algorithm tailored for structured GWAS data. In terms of risk prediction, we show empirically on several GWAS datasets that the proposed T-Trees method significantly outperforms both the original Random Forest algorithm and standard linear models, thereby suggesting the actual existence of multivariate non-linear effects due to the combinations of several SNPs. We also demonstrate that variable importances as derived from our method can help identify relevant loci. Finally, we highlight the strong impact that quality control procedures may have, both in terms of predictive power and loci identification. Variable importance results and T-Trees source code are all available at www.montefiore.ulg.ac.be/~botta/ttrees/ and github.com/0asa/TTree-source respectively.

Multivariate generalized multifactor dimensionality reduction to detect gene-gene interactions

Background

Recently, one of the greatest challenges in genome-wide association studies is to detect gene-gene and/or gene-environment interactions for common complex human diseases. Ritchie et al. (2001) proposed multifactor dimensionality reduction (MDR) method for interaction analysis. MDR is a combinatorial approach to reduce multi-locus genotypes into high-risk and low-risk groups. Although MDR has been widely used for case-control studies with binary phenotypes, several extensions have been proposed. One of these methods, a generalized MDR (GMDR) proposed by Lou et al. (2007), allows adjusting for covariates and applying to both dichotomous and continuous phenotypes. GMDR uses the residual score of a generalized linear model of phenotypes to assign either high-risk or low-risk group, while MDR uses the ratio of cases to controls.

Methods

In this study, we propose multivariate GMDR, an extension of GMDR for multivariate phenotypes. Jointly analysing correlated multivariate phenotypes may have more power to detect susceptible genes and gene-gene interactions. We construct generalized estimating equations (GEE) with multivariate phenotypes to extend generalized linear models. Using the score vectors from GEE we discriminate high-risk from low-risk groups. We applied the multivariate GMDR method to the blood pressure data of the 7,546 subjects from the Korean Association Resource study: systolic blood pressure (SBP) and diastolic blood pressure (DBP). We compare the results of multivariate GMDR for SBP and DBP to the results from separate univariate GMDR for SBP and DBP, respectively. We also applied the multivariate GMDR method to the repeatedly measured hypertension status from 5,466 subjects and compared its result with those of univariate GMDR at each time point.

Results

Results from the univariate GMDR and multivariate GMDR in two-locus model with both blood pressures and hypertension phenotypes indicate best combinations of SNPs whose interaction has significant association with risk for high blood pressures or hypertension. Although the test balanced accuracy (BA) of multivariate analysis was not always greater than that of univariate analysis, the multivariate BAs were more stable with smaller standard deviations.

Conclusions

In this study, we have developed multivariate GMDR method using GEE approach. It is useful to use multivariate GMDR with correlated multiple phenotypes of interests.

Review: toxicometabolomics

Metabolomics use in toxicology is rapidly increasing, particularly owing to advances in mass spectroscopy, which is widely used in the life sciences for phenotyping disease states. Toxicology has the advantage of having the disease agent, the toxicant, available for experimental induction of metabolomics changes monitored over time and dose. This review summarizes the different technologies employed and gives examples of their use in various areas of toxicology. A prominent use of metabolomics is the identification of signatures of toxicity - patterns of metabolite changes predictive of a hazard manifestation. Increasingly, such signatures indicative of a certain hazard manifestation are identified, suggesting that certain modes of action result in specific derangements of the metabolism. This might enable the deduction of underlying pathways of toxicity, which, in their entirety, form the Human Toxome, a key concept for implementing the vision of Toxicity Testing for the 21st century. This review summarizes the current state of metabolomics technologies and principles, their uses in toxicology and gives a thorough overview on metabolomics bioinformatics, pathway identification and quality assurance. In addition, this review lays out the prospects for further metabolomics application also in a regulatory context.

Using information interaction to discover epistatic effects in complex diseases

It is widely agreed that complex diseases are typically caused by the joint effects of multiple instead of a single genetic variation. These genetic variations may show stronger effects when considered together than when considered individually, a phenomenon known as epistasis or multilocus interaction. In this work, we explore the applicability of information interaction to discover pairwise epistatic effects related with complex diseases. We start by showing that traditional approaches such as classification methods or greedy feature selection methods (such as the Fleuret method) do not perform well on this problem. We then compare our information interaction method with BEAM and SNPHarvester in artificial datasets simulating epistatic interactions and show that our method is more powerful to detect pairwise epistatic interactions than its competitors. We show results of the application of information interaction method to the WTCCC breast cancer dataset. Our results are validated using permutation tests. We were able to find 89 statistically significant pairwise interactions with a p-value lower than . Even though many recent algorithms have been designed to find epistasis with low marginals, we observed that all (except one) of the SNPs involved in statistically significant interactions have moderate or high marginals. We also report that the interactions found in this work were not present in gene-gene interaction network STRING.

The association of genetic markers for type 2 diabetes with prediabetic status - cross-sectional data of a diabetes prevention trial

OBJECTIVE

To investigate the association of risk alleles for type 2 diabetes with prediabetes accounting for age, anthropometry, inflammatory markers and lifestyle habits.

DESIGN

Cross-sectional study of 129 men and 157 women of medium-sized companies in northern Germany in the Delay of Impaired Glucose Tolerance by a Healthy Lifestyle Trial (DELIGHT).

METHODS

Besides established risk factors, 41 single nucleotide polymorphisms (SNPs) that have previously been found to be associated with type 2 diabetes were analyzed. As a nonparametric test a random forest approach was used that allows processing of a large number of predictors. Variables with the highest impact were entered into a multivariate logistic regression model to estimate their association with prediabetes.

RESULTS

Individuals with prediabetes were characterized by a slightly, but significantly higher number of type 2 diabetes risk alleles (42.5±4.1 vs. 41.3±4.1, p = 0.013). After adjustment for age and waist circumference 6 SNPs with the highest impact in the random forest analysis were associated with risk for prediabetes in a logistic regression model. At least 5 of these SNPs were positively related to prediabetic status (odds ratio for prediabetes 1.57 per allele (Cl 1.21-2.10, p = 0.001)).

CONCLUSIONS

This explorative analysis of data of DELIGHT demonstrates that at least 6 out of 41 genetic variants characteristic of individuals with type 2 diabetes may also be associated with prediabetes. Accumulation of these risk alleles may markedly increase the risk for prediabetes. However, prospective studies are required to corroborate these findings and to demonstrate the predictive value of these genetic variants for the risk to develop prediabetes.

Summarizing techniques that combine three non-parametric scores to detect disease-associated 2-way SNP-SNP interactions

Identifying susceptibility genes that influence complex diseases is extremely difficult because loci often influence the disease state through genetic interactions. Numerous approaches to detect disease-associated SNP-SNP interactions have been developed, but none consistently generates high-quality results under different disease scenarios. Using summarizing techniques to combine a number of existing methods may provide a solution to this problem. Here we used three popular non-parametric methods-Gini, absolute probability difference (APD), and entropy-to develop two novel summary scores, namely principle component score (PCS) and Z-sum score (ZSS), with which to predict disease-associated genetic interactions. We used a simulation study to compare performance of the non-parametric scores, the summary scores, the scaled-sum score (SSS; used in polymorphism interaction analysis (PIA)), and the multifactor dimensionality reduction (MDR). The non-parametric methods achieved high power, but no non-parametric method outperformed all others under a variety of epistatic scenarios. PCS and ZSS, however, outperformed MDR. PCS, ZSS and SSS displayed controlled type-I-errors (<0.05) compared to GS, APDS, ES (>0.05). A real data study using the genetic-analysis-workshop 16 (GAW 16) rheumatoid arthritis dataset identified a number of interesting SNP-SNP interactions.

An omnibus permutation test on ensembles of two-locus analyses can detect pure epistasis and genetic heterogeneity in genome-wide association studies

This article presents the ability of an omnibus permutation test on ensembles of two-locus analyses (2LOmb) to detect pure epistasis in the presence of genetic heterogeneity. The performance of 2LOmb is evaluated in various simulation scenarios covering two independent causes of complex disease where each cause is governed by a purely epistatic interaction. Different scenarios are set up by varying the number of available single nucleotide polymorphisms (SNPs) in data, number of causative SNPs and ratio of case samples from two affected groups. The simulation results indicate that 2LOmb outperforms multifactor dimensionality reduction (MDR) and random forest (RF) techniques in terms of a low number of output SNPs and a high number of correctly-identified causative SNPs. Moreover, 2LOmb is capable of identifying the number of independent interactions in tractable computational time and can be used in genome-wide association studies. 2LOmb is subsequently applied to a type 1 diabetes mellitus (T1D) data set, which is collected from a UK population by the Wellcome Trust Case Control Consortium (WTCCC). After screening for SNPs that locate within or near genes and exhibit no marginal single-locus effects, the T1D data set is reduced to 95,991 SNPs from 12,146 genes. The 2LOmb search in the reduced T1D data set reveals that 12 SNPs, which can be divided into two independent sets, are associated with the disease. The first SNP set consists of three SNPs from MUC21 (mucin 21, cell surface associated), three SNPs from MUC22 (mucin 22), two SNPs from PSORS1C1 (psoriasis susceptibility 1 candidate 1) and one SNP from TCF19 (transcription factor 19). A four-locus interaction between these four genes is also detected. The second SNP set consists of three SNPs from ATAD1 (ATPase family, AAA domain containing 1). Overall, the findings indicate the detection of pure epistasis in the presence of genetic heterogeneity and provide an alternative explanation for the aetiology of T1D in the UK population.

Transcriptome classification reveals molecular subtypes in psoriasis

BACKGROUND

Psoriasis is an immune-mediated disease characterised by chronically elevated pro-inflammatory cytokine levels, leading to aberrant keratinocyte proliferation and differentiation. Although certain clinical phenotypes, such as plaque psoriasis, are well defined, it is currently unclear whether there are molecular subtypes that might impact on prognosis or treatment outcomes.

RESULTS

We present a pipeline for patient stratification through a comprehensive analysis of gene expression in paired lesional and non-lesional psoriatic tissue samples, compared with controls, to establish differences in RNA expression patterns across all tissue types. Ensembles of decision tree predictors were employed to cluster psoriatic samples on the basis of gene expression patterns and reveal gene expression signatures that best discriminate molecular disease subtypes. This multi-stage procedure was applied to several published psoriasis studies and a comparison of gene expression patterns across datasets was performed.

CONCLUSION

Overall, classification of psoriasis gene expression patterns revealed distinct molecular sub-groups within the clinical phenotype of plaque psoriasis. Enrichment for TGFb and ErbB signaling pathways, noted in one of the two psoriasis subgroups, suggested that this group may be more amenable to therapies targeting these pathways. Our study highlights the potential biological relevance of using ensemble decision tree predictors to determine molecular disease subtypes, in what may initially appear to be a homogenous clinical group. The R code used in this paper is available upon request.

A survey of application: genomics and genetic programming, a new frontier

The aim of this paper is to provide an introduction to the rapidly developing field of genetic programming (GP). Particular emphasis is placed on the application of GP to genomics. First, the basic methodology of GP is introduced. This is followed by a review of applications in the areas of gene network inference, gene expression data analysis, SNP analysis, epistasis analysis and gene annotation. Finally this paper concluded by suggesting potential avenues of possible future research on genetic programming, opportunities to extend the technique, and areas for possible practical applications.

Lessons from genome-wide association studies findings in Alzheimer's disease

Alzheimer's disease (AD) is the most common neurodegenerative disorder with a complex genetic background. Recent genome-wide association studies (GWAS) have placed important new contributors into the genetic framework of early- and late-onset forms of this dementia. Besides confirming the major role of classic allelic variants (e.g. apolipoprotein E) in the development of AD, GWAS have thus far implicated over 20 single nucleotide polymorphisms in AD. In this review, we summarize the findings of 16 AD-based GWAS performed to date whose public registries are available at the National Human Genome Research Institute, with an emphasis on understanding whether the polymorphic markers under consideration support functional implications to the pathophysiological role of the major genetic risk factors unraveled by GWAS.

A hybrid framework for genome wide epistasis discovery

A hybrid framework integrating Random Forest and Logistic Regression is proposed and implemented for genome-wide epistasis study. The two-stage approach first uses random forest model to capture a pool of epistasis-prone single nucleotide polymorphisms (SNPs), followed by using logistic regression to identify the significant pair-wise epistasis SNPs. We tested the proposed framework on data obtained from Singapore Malay Eye Study (SiMES), in which, 3280 subjects were genotyped on Illumina 610 quad arrays and optic nerve parameters were measured in ocular examination. Case-control data set is labeled by choosing the high/low end of vertical Cup-to-Disc ratio (vCDR) values which is a measure of optic nerve degeneration. Our method identified 230 pairs of interacting SNPs with P-values below 5 × 10(-8). A preliminary search identified a protein interaction network at a high confidence score of 0.9. The proteins are known to participate in the WNT pathway with involvement in the survival and differentiation of the retina ganglion cells, inferring a strong association with vCDR. The experimental results demonstrate that the proposed framework is valid and efficient for large scale epistatsis study.

Improved Bevirimat resistance prediction by combination of structural and sequence-based classifiers

BACKGROUND

Maturation inhibitors such as Bevirimat are a new class of antiretroviral drugs that hamper the cleavage of HIV-1 proteins into their functional active forms. They bind to these preproteins and inhibit their cleavage by the HIV-1 protease, resulting in non-functional virus particles. Nevertheless, there exist mutations in this region leading to resistance against Bevirimat. Highly specific and accurate tools to predict resistance to maturation inhibitors can help to identify patients, who might benefit from the usage of these new drugs.

RESULTS

We tested several methods to improve Bevirimat resistance prediction in HIV-1. It turned out that combining structural and sequence-based information in classifier ensembles led to accurate and reliable predictions. Moreover, we were able to identify the most crucial regions for Bevirimat resistance computationally, which are in line with experimental results from other studies.

CONCLUSIONS

Our analysis demonstrated the use of machine learning techniques to predict HIV-1 resistance against maturation inhibitors such as Bevirimat. New maturation inhibitors are already under development and might enlarge the arsenal of antiretroviral drugs in the future. Thus, accurate prediction tools are very useful to enable a personalized therapy.

Methods for identifying SNP interactions: a review on variations of Logic Regression, Random Forest and Bayesian logistic regression

Due to advancements in computational ability, enhanced technology and a reduction in the price of genotyping, more data are being generated for understanding genetic associations with diseases and disorders. However, with the availability of large data sets comes the inherent challenges of new methods of statistical analysis and modeling. Considering a complex phenotype may be the effect of a combination of multiple loci, various statistical methods have been developed for identifying genetic epistasis effects. Among these methods, logic regression (LR) is an intriguing approach incorporating tree-like structures. Various methods have built on the original LR to improve different aspects of the model. In this study, we review four variations of LR, namely Logic Feature Selection, Monte Carlo Logic Regression, Genetic Programming for Association Studies, and Modified Logic Regression-Gene Expression Programming, and investigate the performance of each method using simulated and real genotype data. We contrast these with another tree-like approach, namely Random Forests, and a Bayesian logistic regression with stochastic search variable selection.

The challenges for molecular nutrition research 1: linking genotype to healthy nutrition

Nutrition science finds itself at a major crossroad. On the one hand we can continue the current path, which has resulted in some substantial advances, but also many conflicting messages which impair the trust of the general population, especially those who are motivated to improve their health through diet. The other road is uncharted and is being built over the many exciting new developments in life sciences. This new era of nutrition recognizes the complex relation between the health of the individual, its genome, and the life-long dietary exposure, and has lead to the realisation that nutrition is essentially a gene-environment interaction science. This review on the relation between genotype, diet and health is the first of a series dealing with the major challenges in molecular nutrition, analyzing the foundations of nutrition research. With the unravelling of the human genome and the linking of its variability to a multitude of phenotypes from "healthy" to an enormously complex range of predispositions, the dietary modulation of these propensities has become an area of active research. Classical genetic approaches applied so far in medical genetics have steered away from incorporating dietary effects in their models and paradoxically, most genetic studies analyzing diet-associated phenotypes and diseases simply ignore diet. Yet, a modest but increasing number of studies are accounting for diet as a modulator of genetic associations. These range from observational cohorts to intervention studies with prospectively selected genotypes. New statistical and bioinformatics approaches are becoming available to aid in design and evaluation of these studies. This review discusses the various approaches used and provides concrete recommendations for future research.

Learning genetic epistasis using Bayesian network scoring criteria

Background

Gene-gene epistatic interactions likely play an important role in the genetic basis of many common diseases. Recently, machine-learning and data mining methods have been developed for learning epistatic relationships from data. A well-known combinatorial method that has been successfully applied for detecting epistasis is Multifactor Dimensionality Reduction (MDR). Jiang et al. created a combinatorial epistasis learning method called BNMBL to learn Bayesian network (BN) epistatic models. They compared BNMBL to MDR using simulated data sets. Each of these data sets was generated from a model that associates two SNPs with a disease and includes 18 unrelated SNPs. For each data set, BNMBL and MDR were used to score all 2-SNP models, and BNMBL learned significantly more correct models. In real data sets, we ordinarily do not know the number of SNPs that influence phenotype. BNMBL may not perform as well if we also scored models containing more than two SNPs. Furthermore, a number of other BN scoring criteria have been developed. They may detect epistatic interactions even better than BNMBL.

Although BNs are a promising tool for learning epistatic relationships from data, we cannot confidently use them in this domain until we determine which scoring criteria work best or even well when we try learning the correct model without knowledge of the number of SNPs in that model.

Results

We evaluated the performance of 22 BN scoring criteria using 28,000 simulated data sets and a real Alzheimer's GWAS data set. Our results were surprising in that the Bayesian scoring criterion with large values of a hyperparameter called α performed best. This score performed better than other BN scoring criteria and MDR at recall using simulated data sets, at detecting the hardest-to-detect models using simulated data sets, and at substantiating previous results using the real Alzheimer's data set.

Conclusions

We conclude that representing epistatic interactions using BN models and scoring them using a BN scoring criterion holds promise for identifying epistatic genetic variants in data. In particular, the Bayesian scoring criterion with large values of a hyperparameter α appears more promising than a number of alternatives.

Model-Based Multifactor Dimensionality Reduction to detect epistasis for quantitative traits in the presence of error-free and noisy data

Detecting gene-gene interactions or epistasis in studies of human complex diseases is a big challenge in the area of epidemiology. To address this problem, several methods have been developed, mainly in the context of data dimensionality reduction. One of these methods, Model-Based Multifactor Dimensionality Reduction, has so far mainly been applied to case-control studies. In this study, we evaluate the power of Model-Based Multifactor Dimensionality Reduction for quantitative traits to detect gene-gene interactions (epistasis) in the presence of error-free and noisy data. Considered sources of error are genotyping errors, missing genotypes, phenotypic mixtures and genetic heterogeneity. Our simulation study encompasses a variety of settings with varying minor allele frequencies and genetic variance for different epistasis models. On each simulated data, we have performed Model-Based Multifactor Dimensionality Reduction in two ways: with and without adjustment for main effects of (known) functional SNPs. In line with binary trait counterparts, our simulations show that the power is lowest in the presence of phenotypic mixtures or genetic heterogeneity compared to scenarios with missing genotypes or genotyping errors. In addition, empirical power estimates reduce even further with main effects corrections, but at the same time, false-positive percentages are reduced as well. In conclusion, phenotypic mixtures and genetic heterogeneity remain challenging for epistasis detection, and careful thought must be given to the way important lower-order effects are accounted for in the analysis.

A principal components-based clustering method to identify variants associated with complex traits

BACKGROUND

Multivariate methods ranging from joint SNP to principal components analysis (PCA) have been developed for testing multiple markers in a region for association with disease and disease-related traits. However, these methods suffer from low power and/or the inability to identify the subset of markers contributing to evidence for association under various scenarios.

METHODS

We introduce orthoblique principal components-based clustering (OPCC) as an alternative approach to identify specific subsets of markers showing association with a quantitative outcome of interest. We demonstrate the utility of OPCC using simulation studies and an example from the literature on type 2 diabetes.

RESULTS

Compared to traditional methods, OPCC has similar or improved power under various scenarios of linkage disequilibrium structure and genotype availability. Most importantly, our simulations show how OPCC accurately parses large numbers of markers to a subset containing the causal variant or its proxy.

CONCLUSION

OPCC is a powerful and efficient data reduction method for detecting associations between gene variants and disease-related traits. Unlike alternative methodologies, OPCC has the ability to isolate the effect of causal SNP(s) from among large sets of markers in a candidate region. Therefore, OPCC is an improvement over PCA for testing multiple SNP associations with phenotypes of interest.

Practical and theoretical considerations in study design for detecting gene-gene interactions using MDR and GMDR approaches

Detection of interacting risk factors for complex traits is challenging. The choice of an appropriate method, sample size, and allocation of cases and controls are serious concerns. To provide empirical guidelines for planning such studies and data analyses, we investigated the performance of the multifactor dimensionality reduction (MDR) and generalized MDR (GMDR) methods under various experimental scenarios. We developed the mathematical expectation of accuracy and used it as an indicator parameter to perform a gene-gene interaction study. We then examined the statistical power of GMDR and MDR within the plausible range of accuracy (0.50∼0.65) reported in the literature. The GMDR with covariate adjustment had a power of>80% in a case-control design with a sample size of≥2000, with theoretical accuracy ranging from 0.56 to 0.62. However, when the accuracy was<0.56, a sample size of≥4000 was required to have sufficient power. In our simulations, the GMDR outperformed the MDR under all models with accuracy ranging from 0.56∼0.62 for a sample size of 1000–2000. However, the two methods performed similarly when the accuracy was outside this range or the sample was significantly larger. We conclude that with adjustment of a covariate, GMDR performs better than MDR and a sample size of 1000∼2000 is reasonably large for detecting gene-gene interactions in the range of effect size reported by the current literature; whereas larger sample size is required for more subtle interactions with accuracy<0.56.

Identifying genetic interactions in genome-wide data using Bayesian networks

It is believed that interactions among genes (epistasis) may play an important role in susceptibility to common diseases (Moore and Williams [2002]. Ann Med 34:88-95; Ritchie et al. [2001]. Am J Hum Genet 69:138-147). To study the underlying genetic variants of diseases, genome-wide association studies (GWAS) that simultaneously assay several hundreds of thousands of SNPs are being increasingly used. Often, the data from these studies are analyzed with single-locus methods (Lambert et al. [2009]. Nat Genet 41:1094-1099; Reiman et al. [2007]. Neuron 54:713-720). However, epistatic interactions may not be easily detected with single-locus methods (Marchini et al. [2005]. Nat Genet 37:413-417). As a result, both parametric and nonparametric multi-locus methods have been developed to detect such interactions (Heidema et al. [2006]. BMC Genet 7:23). However, efficiently analyzing epistasis using high-dimensional genome-wide data remains a crucial challenge. We develop a method based on Bayesian networks and the minimum description length principle for detecting epistatic interactions. We compare its ability to detect gene-gene interactions and its efficiency to that of the combinatorial method multifactor dimensionality reduction (MDR) using 28,000 simulated data sets generated from 70 different genetic models We further apply the method to over 300,000 SNPs obtained from a GWAS involving late onset Alzheimer's disease (LOAD). Our method outperforms MDR and we substantiate previous results indicating that the GAB2 gene is associated with LOAD. To our knowledge, this is the first successful model-based epistatic analysis using a high-dimensional genome-wide data set.

Xenobiotic metabolizing and transporter genes: gene–gene interactions in schizophrenia and related disorders

Aims: In this study we explored possible epistasis between CYP2D6 (*3, *4, *5, *6 and *1xN), CYP3A5 (*3), CYP1A2 (*1C and *1F) and ABCB1 (G2677T) in schizophrenia and related disorders. Materials & methods: A total of 344 patients diagnosed with schizophrenia and related disorders, and 484 healthy controls participated in the present study. We analyzed gene–gene interactions by multifactor dimensionality reduction. Results: A four-way model including ABCB1 G2677T, CYP3A5*3, CYP1A2*1F and CYP2D6*4 variants had the best overall performances (accuracy: 0.573) and a crossvalidation consistency of 10/10 (permutation testing p < 0.004). Conclusion: Our results suggest a significant involvement of CYPs and transporters in brain metabolism and homeostasis, and provide evidence of gene–gene interactions among xenobiotic metabolizing and transporter genes in the context of schizophrenia.

Importance measures for epistatic interactions in case-parent trios

Ensemble methods (such as Bagging and Random Forests) take advantage of unstable base learners (such as decision trees) to improve predictions, and offer measures of variable importance useful for variable selection. LogicFS has been proposed as such an ensemble learner for case-control studies when interactions of single nucleotide polymorphisms (SNPs) are of particular interest. LogicFS uses bootstrap samples of the data and employs the Boolean trees derived in logic regression as base learners to create ensembles of models that allow for the quantification of the contributions of epistatic interactions to the disease risk. In this article, we propose an extension of logicFS suitable for case-parent trio data, and derive an additional importance measure that is much less influenced by linkage disequilibrium between SNPs than the measure originally used in logicFS. We illustrate the performance of the novel procedure in simulation studies and in a case study of 461 case-parent trios with autistic children.

A computationally fast measure of epistasis for 2 SNPs and a categorical phenotype

Complex diseases may be caused by interactions or combined effects between multiple genetic and environmental factors. One of the main limitations of testing for interaction between genetic loci in large whole genome studies is the high computational cost of performing such analyses. In this study a new methodology for interaction testing (commonly referred to in genetics as the epistatic effect) between two single nucleotide polymorphisms (SNPs) and a categorical phenotype is presented. It is shown that it provides reasonable approximations with a significantly shorter run time. The proposed measure based on the Pearson's chi-square additive property is compared to fitting a logistic regression model on a randomly selected subset of 218 SNP loci from a study that included 550,000 SNPs). For each possible pair of SNPs a chi-square test for the epistatic effect on case-control status is estimated by fitting a logistic regression model, and compared to the results of the proposed method. Results indicate strong agreement (Pearson's correlation r>0.95) while the proposed method is found to be 20 times faster. This provides a significant pragmatic advantage for the proposed method since the number of tests for epistasis can now be increased by a factor of 20 while the computational cost remains the same.

A fast algorithm for learning epistatic genomic relationships

Genetic epidemiologists strive to determine the genetic profile of diseases. Epistasis is the interaction between two or more genes to affect phenotype. Due to the often non-linearity of the interaction, it is difficult to detect statistical patterns of epistasis. Combinatorial methods for detecting epistasis investigate a subset of combinations of genes without employing a search strategy. Therefore, they do not scale to handling the high-dimensional data found in genome-wide association studies (GWAS). We represent genome-phenome interactions using a Bayesian network rule, which is a specialized Bayesian network. We develop an efficient search algorithm to learn from data a high scoring rule that may contain two or more interacting genes. Our experimental results using synthetic data indicate that this algorithm detects interacting genes as well as a Bayesian network combinatorial method, and it is much faster. Our results also indicate that the algorithm can successfully learn genome-phenome relationships using a real GWAS dataset.