Learning Bayesian Networks Research Articles

Characterization of Nucleoside Reverse Transcriptase Inhibitor-Associated Mutations in the RNase H Region of HIV-1 Subtype C Infected Individuals.

The South African national treatment programme includes nucleoside reverse transcriptase inhibitors (NRTIs) in both first and second line highly active antiretroviral therapy regimens. Mutations in the RNase H domain have been associated with resistance to NRTIs but primarily in HIV-1 subtype B studies. Here, we investigated the prevalence and association of RNase H mutations with NRTI resistance in sequences from HIV-1 subtype C infected individuals. RNase H sequences from 112 NRTI treated but virologically failing individuals and 28 antiretroviral therapy (ART)-naive individuals were generated and analysed. In addition, sequences from 359 subtype C ART-naive sequences were downloaded from Los Alamos database to give a total of 387 sequences from ART-naive individuals for the analysis. Fisher’s exact test was used to identify mutations and Bayesian network learning was applied to identify novel NRTI resistance mutation pathways in RNase H domain. The mutations A435L, S468A, T470S, L484I, A508S, Q509L, L517I, Q524E and E529D were more prevalent in sequences from treatment-experienced compared to antiretroviral treatment naive individuals, however, only the E529D mutation remained significant after correction for multiple comparison. Our findings suggest a potential interaction between E529D and NRTI-treatment; however, site-directed mutagenesis is needed to understand the impact of this RNase H mutation.

A targeted Bayesian network learning for classification

A targeted Bayesian network learning (TBNL) method is proposed to account for a classification objective during the learning stage of the network model. The TBNL approximates the expected conditional probability distribution of the class variable. It effectively manages the trade-off between the classification accuracy and the model complexity by using a discriminative approach, constrained by information theory measurements. The proposed approach also provides a mechanism for maximizing the accuracy via a Pareto frontier over a complexity–accuracy plane, in cases of missing data in the data-sets. A comparative study over a set of classification problems shows the competitiveness of the TBNL mainly with respect to other graphical classifiers.

On conditional truncated densities Bayesian networks

The majority of Bayesian networks learning and inference algorithms rely on the assumption that all random variables are discrete, which is not necessarily the case in real-world problems. In situations where some variables are continuous, a trade-off between the expressive power of the model and the computational complexity of inference has to be done: on one hand, conditional Gaussian models are computationally efficient but they lack expressive power; on the other hand, mixtures of exponentials (MTE), basis functions (MTBF) or polynomials (MOP) are expressive but this comes at the expense of tractability. In this paper, we introduce an alternative model called a ctdBN that lies in between. It is composed of a “discrete” Bayesian network (BN) combined with a set of univariate conditional truncated densities modeling the uncertainty over the continuous random variables given their discrete counterpart resulting from a discretization process. We prove that ctdBNs can approximate (arbitrarily well) any Lipschitz mixed probability distribution. They can therefore be exploited in many practical situations. An efficient inference algorithm is also provided and its computational complexity justifies theoretically why inference computation times in ctdBNs are very close to those in discrete BNs. Experiments confirm the tractability of the model and highlight its expressive power, notably by comparing it with BNs on classification problems and with MTEs and MOPs on marginal distributions estimations.

A machine learning bayesian network for refrigerant charge faults of variable refrigerant flow air conditioning system

An intelligent fault diagnosis network for variable refrigerant flow air conditioning system is proposed in this study. The network is developed under the foundation of bayesian belief network theory, which comprises two main elements: the structure and parameters. The structure obtained by machine learning and experts’ experiences illustrates the relationships among faults and physical variables from the qualitative prospective, and its parameters (including prior probability distribution and conditional distribution) describe the uncertainty between them quantitatively. Once the structure and parameters are determined, the posterior probability distribution which can be used to complete fault diagnosis and isolation will be calculated by some algorithms. In comparison with other fault diagnosis approaches, the proposed approach can make full use of performance information. Moreover, it is more reasonable and precise to express the relationship between faults and variables rather than Boolean variables. Evaluation was conducted on a variable refrigerant flow air conditioning system, which demonstrated that this strategy is effective and efficient.

Combination of expert decision and learned based Bayesian Networks for multi-scale mechanical analysis of timber elements

The use of Bayesian Networks allows to organize and correlate information gathered from different sources and its optimization may incorporate restrictions adjusting the network based on expert knowledge and network operativeness, in such a way that it may satisfactorily represent a given domain. The main goal of this paper is to study if an optimized learned Bayesian Network may be used as a prior structure for an expert based network of an engineering structural material analysis. The methodology is applied to a database of results from an experimental campaign that focused on the mechanical characterization of timber elements recovered from an early 20th century building. To that study case it is evidenced that through a suitable combination of model averaging and supervision steps it is possible to achieve robust and reliable models to underpin the causal structure of a typical multi-scale timber analysis.

Learning Bayesian network parameters from small data sets: A further constrained qualitatively maximum a posteriori method

To improve the learning accuracy of the parameters in a Bayesian network from a small data set, domain knowledge is normally incorporated into the learning process as parameter constraints. MAP-based (Maximum a Posteriori) methods that utilize both sample data and domain knowledge have been well studied in the literature. Among all the MAP-based methods, the QMAP (Qualitatively Maximum a Posteriori) method exhibits the best learning performance. However, when the data is insufficient, the estimation given by the QMAP often fails to satisfy all the convex parameter constraints, and this has made the overall QMAP estimation unreliable. To ensure that QMAP estimation does not violate any given parameter constraints and to further improve the learning accuracy, a FC-QMAP (Further Constrained Qualitatively Maximum a Posteriori) algorithm is proposed in this paper. The algorithm regulates QMAP estimation by replacing data estimation with a further constrained estimation via convex optimization. Experiments and theoretical analysis show that the proposed algorithm outperforms most of the existing parameter learning methods (namely, Maximum Likelihood, Constrained Maximum Likelihood, Maximum Entropy, Constrained Maximum Entropy, Maximum a Posteriori, and Qualitatively Maximum a Posteriori).

Applying Nonparametric Methods to Analyses of Short-Term Fine Particulate Matter Exposure and Hospital Admissions for Cardiovascular Diseases among Older Adults.

Short-term exposure to fine particulate matter (PM2.5) has been associated with increased risks of cardiovascular diseases (CVDs), but whether such associations are supportive of a causal relationship is unclear, and few studies have employed formal causal analysis methods to address this. We employed nonparametric methods to examine the associations between daily concentrations of PM2.5 and hospital admissions (HAs) for CVD among adults aged 75 years and older in Texas, USA. We first quantified the associations in partial dependence plots generated using the random forest approach. We next used a Bayesian network learning algorithm to identify conditional dependencies between CVD HAs of older men and women and several predictor variables. We found that geographic location (county), time (e.g., month and year), and temperature satisfied necessary information conditions for being causes of CVD HAs among older men and women, but daily PM2.5 concentrations did not. We also found that CVD HAs of disjoint subpopulations were strongly predictive of CVD HAs among older men and women, indicating the presence of unmeasured confounders. Our findings from nonparametric analyses do not support PM2.5 as a direct cause of CVD HAs among older adults.

MapReduce Based Parallel Bayesian Network for Manufacturing Quality Control

Increasing complexity of industrial products and manufacturing processes have challenged conventional statistics based quality management approaches in the circumstances of dynamic production. A Bayesian network and big data analytics integrated approach for manufacturing process quality analysis and control is proposed. Based on Hadoop distributed architecture and MapReduce parallel computing model, big volume and variety quality related data generated during the manufacturing process could be dealt with. Artificial intelligent algorithms, including Bayesian network learning, classification and reasoning, are embedded into the Reduce process. Relying on the ability of the Bayesian network in dealing with dynamic and uncertain problem and the parallel computing power of MapReduce, Bayesian network of impact factors on quality are built based on prior probability distribution and modified with posterior probability distribution. A case study on hull segment manufacturing precision management for ship and offshore platform building shows that computing speed accelerates almost directly proportionally to the increase of computing nodes. It is also proved that the proposed model is feasible for locating and reasoning of root causes, forecasting of manufacturing outcome, and intelligent decision for precision problem solving. The integration of bigdata analytics and BN method offers a whole new perspective in manufacturing quality control.

An Implementation of Parallel Bayesian Network Learning

An Implementation of Parallel Bayesian Network Learning

Learning Discrete Bayesian Networks from Continuous Data

Learning Bayesian networks from raw data can help provide insights into the relationships between variables. While real data often contains a mixture of discrete and continuous-valued variables, many Bayesian network structure learning algorithms assume all random variables are discrete. Thus, continuous variables are often discretized when learning a Bayesian network. However, the choice of discretization policy has significant impact on the accuracy, speed, and interpretability of the resulting models. This paper introduces a principled Bayesian discretization method for continuous variables in Bayesian networks with quadratic complexity instead of the cubic complexity of other standard techniques. Empirical demonstrations show that the proposed method is superior to the established minimum description length algorithm. In addition, this paper shows how to incorporate existing methods into the structure learning process to discretize all continuous variables and simultaneously learn Bayesian network structures.

Representing local structure in Bayesian networks by Boolean functions

A number of studies on learning Bayesian networks have emphasized the importance of exploiting regularities in conditional probability distributions, i.e., local structure. In this paper, we encode local structures as linear combinations of Boolean functions. By using Lasso, we can simultaneously estimate the structure and parameters of the networks from limited data. We demonstrate that the method leads to improved performance in terms of structural correctness as well as prediction score even when the local structure in the underlying model is only implicit.

Surveying computational algorithms for identification of miRNA–mRNA regulatory modules

MicroRNAs (miRNAs) are small non-coding RNA that target gene expression at post-transcriptional level. To understand the molecular basis of miRNA regulation, it is essential to identify reliable miRNA target mRNAs and miRNA–mRNA functional networks. Several computational algorithms and tools are widely used for discovering miRNA–mRNA regulatory modules. We have comprehensively reviewed thirty-two tools and databases that were developed over the past decade. Our in-depth survey considers all significant features including underlying methodologies in each tool for basis of miRNA–mRNA functional module discovery. Current miRNA–mRNA functional module identification tools utilizes sequence based prediction, graph theory combined with Bayesian network, probabilistic and rule learning methods. Further we discuss ease of use, availability, and suitability of these methods, and provide an overview of recent computational algorithms used in miRNA–mRNA regulatory module identification.

Exploring resistance pathways for first-generation NS3/4A protease inhibitors boceprevir and telaprevir using Bayesian network learning

Resistance-associated variants (RAVs) have been shown to influence treatment response to direct-acting antivirals (DAAs) and first generation NS3/4A protease inhibitors (PIs) in particular. Interpretation of hepatitis C virus (HCV) genotypic drug resistance remains a challenge, especially in patients who previously failed DAA therapy and need to be retreated with a second DAA based regimen. Bayesian network (BN) learning on HCV sequence data from patients treated with DAAs could provide insight in resistance pathways against PIs for HCV subtypes 1a and 1b, in a similar way as applied before for HIV. The publicly available ‘Rega-BN’ tool chain was developed to study associative analyses for various pathogens. Our first analysis, comparing sequences from PI-naïve and PI-experienced patients, determined that NS3 substitutions R155K and V36M arise with PI-exposure in HCV1a infected patients, and were defined as major and minor resistance-associated variants respectively. NS3 variant 174H was newly identified as potentially related to PI resistance. In a second analysis, NS3 sequences from PI-naïve patients who cleared the virus during PI therapy and from PI-naïve patients who failed PI therapy were compared, showing that NS3 baseline variant 67S predisposes to treatment-failure and variant 72I to treatment success. This approach has the potential to better characterize the role of more RAVs, if sufficient therapy annotated sequence data becomes available in curated public databases. In addition, polymorphisms present in baseline sequences that predispose patients to therapy failure can be identified using this approach.

Effects of low dose ionizing radiation on DNA damage-caused pathways by reverse-phase protein array and Bayesian networks

Ionizing radiation (IR) causing damages to Deoxyribonucleic acid (DNA) constitutes a broad range of base damage and double strand break, and thereby, it induces the operation of relevant signaling pathways such as DNA repair, cell cycle control, and cell apoptosis. The goal of this paper is to study how the exposure to low dose radiation affects the human body by observing the signaling pathway associated with Ataxia Telangiectasia mutated (ATM) using Reverse-Phase Protein Array (RPPA) and isogenic human Ataxia Telangiectasia (A-T) cells under different amounts and durations of IR exposure. In order to verify which proteins could be involved in a DNA damage-caused pathway, only proteins that highly interact with each other under IR are selected by using correlation coefficient. The pathway inference is derived from learning Bayesian networks in combination with prior knowledge such as Protein-Protein Interactions (PPIs) and signaling pathways from well-known databases. Learning Bayesian networks is based on a score and search scheme that provides the highest scored network structure given a score function, and the prior knowledge is included in the score function as a prior probability by using Dempster-Shafer theory (DST). In this way, the inferred network can be more likely to be similar to already discovered pathways and consistent with confirmed PPIs for more reliable inference. The experimental results show which proteins are involved in signaling pathways under IR, how the inferred pathways are different under low and high doses of IR, and how the selected proteins regulate each other in the inferred pathways. As our main contribution, overall results confirm that low dose IR could cause DNA damage and thereby induce and affect related signaling pathways such as apoptosis, cell cycle, and DNA repair.

Data Missing Bayesian Network Parameters Learning Optimization Algorithm Based on EM

The paper proposes an optimization algorithm for the parameter learning of missing data Bayesian networks. Expectation Maximization (EM) algorithm is the common parameter learning algorithms. The maximum likelihood estimation (MLE) and maximum a posterity estimation (MAP) of EM are local estimate rather than global estimate and are not easy to achieve the global optimal. So this paper puts forward a point estimate relative error minimum optimization algorithm which based on EM algorithm (EM-MLE-MAP). Applying the algorithm to the fault diagnosis of the rotor vibration Bayesian network, simulations and experiments show that the improved algorithm has good precision in diagnosing vibration fault when the loss ratio less than 3 percent.

Learning Bayesian network structures under incremental construction curricula

Bayesian networks have been successfully applied to various tasks for probabilistic reasoning and causal modeling. One major challenge in the application of Bayesian networks is to learn the Bayesian network structures from data. In this paper, we take advantage of the idea of curriculum learning and learn Bayesian network structures by stages. At each stage a subnet is learned over a selected subset of the random variables. The selected subset grows with stages and eventually includes all the variables. We show that in our approach each target subnet is closer to the target Bayesian network than any of its predecessors. The experimental results show that our algorithm outperformed the state-of-the-art heuristic approach in learning Bayesian network structures under several different evaluation metrics.

Untangling the role of one-carbon metabolism in colorectal cancer risk: a comprehensive Bayesian network analysis

The role of one-carbon metabolism (1CM), particularly folate, in colorectal cancer (CRC) development has been extensively studied, but with inconclusive results. Given the complexity of 1CM, the conventional approach, investigating components individually, may be insufficient. We used a machine learning-based Bayesian network approach to study, simultaneously, 14 circulating one-carbon metabolites, 17 related single nucleotide polymorphisms (SNPs), and several environmental factors in relation to CRC risk in 613 cases and 1190 controls from the prospective Northern Sweden Health and Disease Study. The estimated networks corresponded largely to known biochemical relationships. Plasma concentrations of folate (direct), vitamin B6 (pyridoxal 5-phosphate) (inverse), and vitamin B2 (riboflavin) (inverse) had the strongest independent associations with CRC risk. Our study demonstrates the importance of incorporating B-vitamins in future studies of 1CM and CRC development, and the usefulness of Bayesian network learning for investigating complex biological systems in relation to disease.

Learning Bayesian Networks with Incomplete Data by Augmentation

We present new algorithms for learning Bayesian networks from data with missing values using a data augmentation approach. An exact Bayesian network learning algorithm is obtained by recasting the problem into a standard Bayesian network learning problem without missing data. As expected, the exact algorithm does not scale to large domains. We build on the exact method to create an approximate algorithm using a hill-climbing technique. This algorithm scales to large domains so long as a suitable standard structure learning method for complete data is available. We perform a wide range of experiments to demonstrate the benefits of learning Bayesian networks with such new approach.

Using targeted Bayesian network learning for suspect identification in communication networks

This paper proposes a machine learning application to identify mobile phone users suspected of involvement in criminal activities. The application characterizes the behavioral patterns of suspect users versus non-suspect users based on usage metadata such as call duration, call distribution, interaction time preferences and text-to-call ratios while avoiding any access to the content of calls or messages. The application is based on targeted Bayesian network learning method. It generates a graphical network that can be used by domain experts to gain intuitive insights about the key features that can help identify suspect users. The method enables experts to manage the trade-off between model complexity and accuracy using information theory metrics. Unlike other graphical Bayesian classifiers, the proposed application accomplishes the task required of a security company, namely an accurate suspect identification rate (recall) of at least 50% with no more than a 1% false identification rate. The targeted Bayesian network learning method is also used for additional tasks such as anomaly detection, distinction between “relevant” and “irrelevant” anomalies, and for associating anonymous telephone numbers with existing users by matching behavioral patterns.

Classification of X-Ray images of shipping containers

Smuggling has long played an important role in the inefficiency of economies. To secure the borders against this illegal act, X-Ray Inspection Systems are often deployed at the borders and customs. In this paper, we present a new method for classification of shipping containers X-Ray images, produced in the inspection lines. The aim is to improve the matching accuracy of the presented manifest, which lists the claimed contents of the shipping containers, with the real contents of the container. The proposed method is based on utilizing Scale Invariant Feature Transforms (SIFT) feature vectors, Bag of visual words (BOVW) and tree augmented naive Bayes (TAN) approach for classifying containers X-Ray images. The prior research on classification of X-Ray images of shipping containers has focused mostly on working with greedy algorithms such as sliding windows for task of classification. More recent studies introduced filter banks and visual words for extraction of features. The proposed method for the first time considers the salient points and keypoints for the task of feature extraction. In addition, this paper presents a framework using the tree augmented naive Bayes based on the theory of learning Bayesian networks, which is proved to have a significant improvements upon the prior designed systems by considering the correlations among the extracted features. For experimental evaluations, our method is compared with two recently proposed methods on containers X-Ray images categorization. The results show that the proposed method is more accurate and time-efficient in categorization of X-Ray images.