Figure 4 - uploaded by James R Bradford
Content may be subject to copyright.
Source publication
Bayesian networks (BNs) provide a neat and compact representation for expressing joint probability distributions (JPDs) and for inference. They are becoming increasingly important in the biological sciences for the tasks of inferring cellular networks [1], modelling protein signalling pathways [2], systems biology, data integration [3], classificat...
Context in source publication
Context 1
... connection. For example, when a transcription factor y turns on two genes x and z ( Figure 4). As with a serial connection, evidence is transmitted unless the variable in the connection is instantiated: if the expression level of y is unknown, then evidence of the level of x effects the level of z (since they are co-regulated-if x is highly expressed, then the likely level of y may be inferred, which in turn would influence the expression level of z); if y is known, then the level of z depends only on the expression level of y. z is conditionally independent from x. ...
Citations
... The Bayesian network is a prominent predictive modeling framework, where its nodes often represent genes or proteins interconnected by edges denoting relationships. These edges, informed by experimental data, impart a predictive capability to Bayesian computational methods [15,31], capturing both the strength and direction of connections. Despite over two decades of development in mixed Bayesian networks, combining continuous and discrete variables [8], the field still lacks efficient algorithms for identifying appropriate strong junction trees in their moral graphs. ...
In recent years, mixed Bayesian networks have received increasing attention across various fields for probabilistic reasoning. Though many studies have been devoted to propagation computation on strong junction trees for mixed Bayesian networks, few have addressed the construction of appropriate strong junction trees. In this work, we establish a connection between the minimal strong triangulation for marked graphs and the minimal triangulation for star graphs. We further propose a minimal strong triangulation method for the moral graph of mixed Bayesian networks and develop a polynomial-time algorithm to derive a strong junction tree from this minimal strong triangulation. Moreover, we also focus on the propagation computation of all posteriors on this derived strong junction tree. We conducted multiple numerical experiments to evaluate the performance of our proposed method, demonstrating significant improvements in computational efficiency compared to existing approaches. Experimental results indicate that our minimal strong triangulation approach provides a robust framework for efficient probabilistic inference in mixed Bayesian networks.
... To interpret the network relations (i.e., assign a direction of effect) and measure the goodness of model fit for each variable in our DAG, we used the predict() function with the bayes-lw method (Needham et al., 2007). The bayes-lw method performs both causal prediction and noncausal Bayesian inference using Monte Carlo methods. ...
Debates over the scope of environmental impact, life-cycle, and cost-benefit analysis frequently revolve around disagreements on the causal structure of complex sociotechnical systems. Environmental advocates in the United States have claimed that new electrical interties with Canada increase development of Canadian hydroelectric resources, leading to environmental and health impacts associated with new reservoirs. Assertions of such second-order impacts of two recently proposed 9.5 TWh year-1 transborder transmission projects played a role in their suspension. We demonstrate via Bayesian network modeling that development of Canadian hydroelectric resources is stimulated by price signals and domestic demand rather than increased export capacity per se. However, hydropower exports are increasingly arranged via long-term power purchase agreements that may promote new generation in a way that is not easily modeled with publicly available data. Overall, this work suggests lesser consideration of generation-side impacts in permitting transborder transmission infrastructure while highlighting the need for higher resolution data to model the Quebec-New England-New York energy system at the project scale. More broadly, Bayesian analysis can be used to elucidate causal drivers in evolving sociotechnical systems to develop consensus for the scope of impacts to consider in environmental impact, life cycle, and cost-benefit analysis.
... Information from BN models (conveying linear and higher order regulatory relationships among genes accompanying cancer) are easy to interpret due to its graphical nature, thus providing valuable insights into the properties of the data being analysed and giving rise to new models to be produced. Direct causal relationships is being modelled by BN via direct acyclic graph (DAG) with the Causal Markov Assumption been imposed on it (Needham et al., 2007;Friedman et al., 2000). ...
A network of relationships among cancer-related genes can be reconstructed from high-throughput datasets obtained by deoxyribonucleic acid (DNA) micro-array technologies. However, modelling such biological networks is challenged by the nature of data and the complexities of relationships among biological variables such as genes. In this paper, Bayesian networks are applied to predict novel regulatory relationships among genes in cancer from genomic datasets. The performances of the methods were assessed by standard metrics such as sensitivities and specificities. Furthermore, in order to validate and verify the reliability of the new predicted relationships among the genes, some of the results were examined with experimentally confirmed relationships found by previous research. Interestingly, some predicted regulatory relationships were also found in the literature. This enhances confidence in the newly predicted network of regulatory relationships, which could become hypotheses for further research.
... A predictive model using a Bayesian network was constructed from patients with a follow-up period of approximately 5 years, where, in addition to the blood pressure measurements, serum glucose, urine albumin and the rs16998073 genotype were considered for all the patients. The rationale for using this model when compared to other AI models exists in the fact that missing data can be addressed and computed for, which provides it with more potential for clinical applicability [121,122]. ...
The measurement of physiologic pressure helps diagnose and prevent associated health complications. From typical conventional methods to more complicated modalities, such as the estimation of intracranial pressures, numerous invasive and noninvasive tools that provide us with insight into daily physiology and aid in understanding pathology are within our grasp. Currently, our standards for estimating vital pressures, including continuous BP measurements, pulmonary capillary wedge pressures, and hepatic portal gradients, involve the use of invasive modalities. As an emerging field in medical technology, artificial intelligence (AI) has been incorporated into analyzing and predicting patterns of physiologic pressures. AI has been used to construct models that have clinical applicability both in hospital settings and at-home settings for ease of use for patients. Studies applying AI to each of these compartmental pressures were searched and shortlisted for thorough assessment and review. There are several AI-based innovations in noninvasive blood pressure estimation based on imaging, auscultation, oscillometry and wearable technology employing biosignals. The purpose of this review is to provide an in-depth assessment of the involved physiologies, prevailing methodologies and emerging technologies incorporating AI in clinical practice for each type of compartmental pressure measurement. We also bring to the forefront AI-based noninvasive estimation techniques for physiologic pressure based on microwave systems that have promising potential for clinical practice.
... If state A can regulate another state B, an arrow will point to state B from state A. To demonstrate the probabilistic correlation in a Bayesian network, the edges can be thicker to show a higher probability [81]. Unlike Boolean networks, the edges in Bayesian networks cannot form a loop [82]. ...
... Needham et al. [82] inferred Bayesian networks using machine learning to find the parameters in the networks even if the available data is incomplete. The common method for inferring the Bayesian network is shown in Fig. 3. ...
p>This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible.
Dynamic disease pathways are a combination of complex dynamical processes among bio-molecules in a cell that leads to diseases. Network modeling of disease pathways considers disease-related bio-molecules (e.g. DNA, RNA, transcription factors, enzymes, proteins, and metabolites) and their interaction (e.g. DNA methylation, histone modification, alternative splicing, and protein modification) to study disease progression and predict therapeutic responses. These bio-molecules and their interactions are the basic elements in the study of the misregulation in the disease-related gene expression that lead to abnormal cellular responses. Gene regulatory networks, cell signaling networks, and metabolic networks are the three major types of intracellular networks for the study of the cellular responses elicited from extracellular signals. The disease-related cellular responses can be prevented or regulated by designing control strategies to manipulate these extracellular signals. The paper reviews the regulatory mechanisms, the dynamic models, and the control strategies for each intracellular network. The applications, limitations and the prospective for modeling and control are also discussed.</p
... Bayesian Networks (BNs), also known as Belief Networks, and related models such as Directed Acyclic Graphs (DAGs), Structural Equation Models (SEMs), and Dynamic Bayesian Networks (DBNs) are used in a variety of applications (Pourret et al., 2008) like healthcare (Kyrimi et al., 2021), medicine (Arora et al., 2019), natural language processing (Goyal et al., 2008), computational biology (Needham et al., 2007;Chen et al., 2022), robotics (Lazkano et al., 2007;Premebida et al., 2016;Amiri et al., 2022), neuroscience (Bielza and Larrañaga, 2014), and others. Some of the most common tasks in these applications are learning model structure from data (structure learning), estimating model parameters (parameter learning), querying a model for conditional or marginal distributions (probabilistic inference), causal effect estimation between variables (causal inference), and simulating data under various generating processes. ...
Bayesian Networks (BNs) are used in various fields for modeling, prediction, and decision making. pgmpy is a python package that provides a collection of algorithms and tools to work with BNs and related models. It implements algorithms for structure learning, parameter estimation, approximate and exact inference, causal inference, and simulations. These implementations focus on modularity and easy extensibility to allow users to quickly modify/add to existing algorithms, or to implement new algorithms for different use cases. pgmpy is released under the MIT License; the source code is available at: https://github.com/pgmpy/pgmpy, and the documentation at: https://pgmpy.org.
... These neural networks can learn from big data and act as the human brain [23]. Deep learning techniques are applied in many areas: law enforcement, finance, customer service, engineering technology application [98], turbines [99], aero engines [100], bearings [101], etc. Although deep learning models can significantly improve integrity estimation, a recent assessment found that they have not been used in corrosion [102]. ...
One of the biggest problems the maritime industry is currently experiencing is corrosion, resulting in short and long-term damages. Early prediction and proper corrosion monitoring can reduce economic losses. Traditional approaches used in corrosion prediction and detection are time-consuming and challenging to execute in inaccessible areas. Due to these reasons, artificial intelligence-based algorithms have become the most popular tools for researchers. This study discusses state-of-the-art artificial intelligence (AI) methods for marine-related corrosion prediction and detection: (1) predictive maintenance approaches and (2) computer vision and image processing approaches. Furthermore, a brief description of AI is described. The outcomes of this review will bring forward new knowledge about AI and the development of prediction models which can avoid unexpected failures during corrosion detection and maintenance. Moreover, it will expand the understanding of computer vision and image processing approaches for accurately detecting corrosion in images and videos.
... Graphical models can be constructed from discrete or continuous data (Hecker et al., 2009;López-Kleine et al., 2013;Segal et al., 2003), as well as Bayseian network methodologies, which integrate both data types (Needham et al., 2007;Ziebarth & Cui, 2017). But regardless of the particular method of inference, prior information about the investigated gene interactions should be maximized from literature (Hecker et al., 2009). ...
The genetic control of biological traits is not predetermined. Rather, phenotypes emerge through stochastic interactions among genes, their products, and environments. A century since Fisher’s synthesis of Mendelian inheritance with observations from biometric studies, the fundamental challenge toward a population genetics of phenotypic evolution has been to dissect the genetic and nongenetic causes of variation in quantitative traits. Some of the earliest accessible phenotypic variation is expressed in the production and action of messenger RNA (mRNA). Recently, progress in transcriptome profiling has made this biological information technically more accessible. Measurement of its quantity, timing, and sequence has become feasible at genomic scales. However, proper instrumental and theoretical insight is required to make meaning of “gene expression” from count data. Gene expression encompasses cascades of dynamic molecular adjustments that are not directly represented in static measurements. Nonetheless, bulk ribonucleic acid sequencing (RNA-seq) can be performed to generate time-courses sampling relative mRNA abundance across conditions. In turn, time-courses can be analyzed to infer the connectivity of transcriptional networks and to inform experimental design for testing genotype–phenotype hypotheses. Moreover, comparative analyses of predicted gene interactions across environmental, demographic, and developmental conditions can advance functional genomics in nonmodel species by providing a means to assign gene function without the need for well-annotated reference genomes, as well as studies into the effects of epistasis and the conditional strength of selection for individual gene expression. Integrative strategies for linking genetic factors to gene expression variation can therefore empower the study of phenotypic evolution to move beyond phenotypic variance and on to genotype frequencies.
... If state A can regulate another state B, an arrow will point to state B from state A. To demonstrate the probabilistic correlation in a Bayesian network, the edges can be thicker to show a higher probability [81]. Unlike Boolean networks, the edges in Bayesian networks cannot form a loop [82]. ...
... Needham et al. [82] inferred Bayesian networks using machine learning to find the parameters in the networks even if the available data is incomplete. The common method for inferring the Bayesian network is shown in Fig. 3. ...
p>Dynamic disease pathways are a combination of complex dynamical processes among bio-molecules in a cell that leads to diseases. Network modeling of disease pathways considers disease-related bio-molecules (e.g. DNA, RNA, transcription factors, enzymes, proteins, and metabolites) and their interaction (e.g. DNA methylation, histone modification, alternative splicing, and protein modification) to study disease progression and predict therapeutic responses. These bio-molecules and their interactions are the basic elements in the study of the misregulation in the disease-related gene expression that lead to abnormal cellular responses. Gene regulatory networks, cell signaling networks, and metabolic networks are the three major types of intracellular networks for the study of the cellular responses elicited from extracellular signals. The disease-related cellular responses can be prevented or regulated by designing control strategies to manipulate these extracellular signals. The paper reviews the regulatory mechanisms, the dynamic models, and the control strategies for each intracellular network. The applications, limitations and the prospective for modeling and control are also discussed.</p
... If state A can regulate another state B, an arrow will point to state B from state A. To demonstrate the probabilistic correlation in a Bayesian network, the edges can be thicker to show a higher probability [81]. Unlike Boolean networks, the edges in Bayesian networks cannot form a loop [82]. ...
... Needham et al. [82] inferred Bayesian networks using machine learning to find the parameters in the networks even if the available data is incomplete. The common method for inferring the Bayesian network is shown in Fig. 3. ...
p>This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible.
Dynamic disease pathways are a combination of complex dynamical processes among bio-molecules in a cell that leads to diseases. Network modeling of disease pathways considers disease-related bio-molecules (e.g. DNA, RNA, transcription factors, enzymes, proteins, and metabolites) and their interaction (e.g. DNA methylation, histone modification, alternative splicing, and protein modification) to study disease progression and predict therapeutic responses. These bio-molecules and their interactions are the basic elements in the study of the misregulation in the disease-related gene expression that lead to abnormal cellular responses. Gene regulatory networks, cell signaling networks, and metabolic networks are the three major types of intracellular networks for the study of the cellular responses elicited from extracellular signals. The disease-related cellular responses can be prevented or regulated by designing control strategies to manipulate these extracellular signals. The paper reviews the regulatory mechanisms, the dynamic models, and the control strategies for each intracellular network. The applications, limitations and the prospective for modeling and control are also discussed.</p
