Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Bayesian networks”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 235 records · Page 13Linked to original sources
PubMed2004

Inferring pathways and networks with a Bayesian framework.

Numerous mathematical methods have been adapted and developed to quantitatively reverse engineer biological networks, for example, signal transduction pathways, from experimental micro-array data. Compared with stochastic methods, such as Boolean networks, and deterministic methods, such as thermodynamic or differential equation-based models, Bayesian network analysis has the ability to assess, with scoring metrics, causal relations based on conditional probabilities and thus permit hypothesis testing. The goal of this paper is to illustrate the integration of several Bayesian based techniques into a unified Bayesian framework that can infer hepatocellular networks from metabolic data. Reverse engineering of pathways and networks provides a framework for predictive modeling and hypotheses testing to gain deeper insight into living organisms, disease mechanisms, and targeted therapeutics. Evaluating this methodology initially against the known biochemical network provides confidence in the networks that are uncovered from the experimental data using this framework. From the metabolic data we inferred the known sub-networks, such as the tricarboxylic acid (TCA) and urea cycles. In addition, we combined the relationships learned from the data and our current knowledge of the biological system to postulate several alternative metabolic sub-network models that can predict a particular cellular function, such as intracellular triglyceride accumulation.

Algorithms↗
PubMed2026

Causal assessment of Bayesian gene regulatory networks from single-cell transcriptomics.

Gene regulatory network (GRN) inference is an essential tool for revealing dysregulated relationships between genes in different cell types from single-cell transcriptomic (SCT) data. GRNs based on Bayesian networks (BNs) learned from SCT data can elucidate directed regulatory relationships representing complex disease mechanisms and their interplay through graphical modeling. However, software for learning BNs from SCT data is not widely available, nor is software for evaluating the BNs' structural accuracy in representing causal relationships between genes. Here, we describe the scstruc R package. This package provides a suite of BN structure learning algorithms specifically designed to handle SCT data, to evaluate the resulting networks based on the causal relationships they represent regardless of the availability of established molecular interaction networks, and to compare regulatory relationships between conditions. We demonstrated that scstruc can identify biologically relevant differential regulatory relationships between groups on a per-cell basis.

Bayesian networks↗
PubMed2005

Inferring gene transcriptional modulatory relations: a genetical genomics approach.

Bayesian network modeling is a promising approach to define and evaluate gene expression circuits in diverse tissues and cell types under different experimental conditions. The power and practicality of this approach can be improved by restricting the number of potential interactions among genes and by defining causal relations before evaluating posterior probabilities for billions of networks. A newly developed genetical genomics method that combines transcriptome profiling with complex trait analysis now provides strong constraints on network architecture. This method detects those chromosomal intervals responsible for differences in mRNA expression using quantitative trait locus (QTL) mapping. We have developed an efficient Bayesian approach that exploits the genetical genomics method to focus computational effort on the most plausible gene modulatory networks. We exploit a dense marker map for a genetic reference population (GRP) that consists of 32 BXD strains of mice made by intercrossing two progenitor strains--C57BL/6J and DBA/2J. These progenitors differ at approximately 1.3 million known single nucleotide polymorphisms (SNPs), all of which can be exploited to estimate the probability that a gene contains functional polymorphisms that segregate within the GRP. We constructed 66 candidate networks that include all the candidate modulator genes located in the 209 statistically significant trans-acting QTL regions. SNPs that distinguish between the two progenitor strains were used to further winnow the list of candidate modulators. Bayesian network was then used to identify the genetic modulatory relations that best explain the microarray data.

Algorithms↗
PubMed2002

Prediction of contact maps by GIOHMMs and recurrent neural networks using lateral propagation from all four cardinal corners.

MOTIVATION: Accurate prediction of protein contact maps is an important step in computational structural proteomics. Because contact maps provide a translation and rotation invariant topological representation of a protein, they can be used as a fundamental intermediary step in protein structure prediction. RESULTS: We develop a new set of flexible machine learning architectures for the prediction of contact maps, as well as other information processing and pattern recognition tasks. The architectures can be viewed as recurrent neural network implemantations of a class of Bayesian networks we call generalized input-output HMMs (GIOHMMs). For the specific case of contact maps, contextual information is propagated laterally through four hidden planes, one for each cardinal corner. We show that these architectures can be trained from examples and yield contact map predictors that outperform previously reported methods. While several extensions and improvements are in progress, the current version can accurately predict 60.5% of contacts at a distance cutoff of 8 A and 45% of distant contacts at 10 A, for proteins of length up to 300.

Algorithms↗
PubMed2005

A comparison of three modelling approaches for quantitative risk assessment using the case study of Salmonella spp. in poultry meat.

A comprehensive review of both the scientific literature and industry practices was undertaken to identify and quantify all sources of contamination throughout the entire poultry meat production chain by Salmonella spp. This information was used to develop a quantitative risk assessment (QRA) model for Salmonella in the production chain from the breeder farm to the chilled carcass. This was subsequently used as the basis on which to compare the merits of three approaches to QRA modelling in such systems. The original model used a Bayesian Network (BN). The second method was a Markov chain Monte Carlo (MCMC) approach, a numerical Bayesian technique which retained a similar network structure but allowed further development, such as the separation of variability and uncertainty. The third method was a more detailed simulation model. The BN responds immediately to changes, such as entering evidence, because it does not use simulation and can propagate information from any point in the network to all others by Bayesian inference. However, it requires all the variables to be discrete, which introduces errors if continuous variables have to be discretized. These errors can accumulate. The MCMC approach does not require discrete variables while retaining some of the properties of the BN model, such as the ability to draw inferences from evidence. Finally, the simulation offers greater flexibility, such as consideration of the individual carcass, but may be more complex to implement as a result and sacrifices the ability to propagate evidence.

Animals↗
PubMed2000

A retrospective evaluation of a data mining approach to aid finding new adverse drug reaction signals in the WHO international database.

BACKGROUND: The detection of new drug safety signals is of growing importance with ever more new drugs becoming available and exposure to medicines increasing. The task of evaluating information relating to safety lies with national agencies and, for international data, with the World Health Organization Programme for International Drug Monitoring. RATIONALE: An established approach for identifying new drug safety signals from the international database of more than 2 million case reports depends upon clinical experts from around the world. With a very large amount of information to evaluate, such an approach is open to human error. To aid the clinical review, we have developed a new signalling process using Bayesian logic, applied to data mining, within a confidence propagation neural network (Bayesian Confidence Propagation Neural Network; BCPNN). Ultimately, this will also allow the evaluation of complex variables. METHODS: The first part of this study tested the predictive value of the BCPNN in new signal detection as compared with reference literature sources (Martindale's Extra Pharmacopoeia in 1993 and July 2000, and the Physicians Desk Reference in July 2000). In the second part of the study, results with the BCPNN method were compared with those of the former signalling procedure. RESULTS: In the study period (the first quarter of 1993) 107 drug-adverse reaction combinations were highlighted as new positive associations by the BCPNN, and referred to new drugs. 15 drug-adverse reaction combinations on new drugs became negative BCPNN associations in the study period. The BCPNN method detected signals with a positive predictive value of 44% and the negative predictive value was 85%. 17 as yet unconfirmed positive associations could not be dismissed with certainty as false positive signals. Of the 10 drug-adverse reaction signals produced by the former signal detection system from data sent out for review during the study period, 6 were also identified by the BCPNN. These 6 associations have all had a more than 10-fold increase of reports and 4 of them have been included in the reference sources. The remaining 4 signals that were not identified by the BCPNN had a small, or no, increase in the number of reports, and are not listed in the reference sources. CONCLUSION: Our evaluation showed that the BCPNN approach had a high and promising predictive value in identifying early signals of new adverse drug reactions.

Algorithms↗
PubMed2026

Molecular targeted therapy in combination with chemotherapy for the treatment of platinum-resistant/refractory ovarian cancer (PROC): a systematic review and network meta-analysis.

BACKGROUND: Although single-agent chemotherapy is the most common approach for treating platinum-resistant or refractory ovarian cancer (PROC), there is growing evidence that combining molecular targeted agents with chemotherapy is beneficial, especially for certain patient groups. However, the most effective combination regimen remains elusive. OBJECTIVES: This Bayesian network meta-analysis (NMA) aims to identify the best combination therapy for PROC. METHODS: Relevant studies were searched in PubMed, EMBASE, Web of Science and the Cochrane Central Register of Controlled Trials from their inception until October 2024. The primary outcomes were overall survival (OS), progression-free survival (PFS) and adverse events (AEs). Statistical analyses were performed using the GEMTC package (1.0-2) and R 4.2.0. This review was registered in PROSPERO (CRD42023428414). RESULTS: Our analysis of 22 randomized controlled trials (RCTs) (n = 3408) demonstrated that chemotherapy combinations with bevacizumab (hazard ratio (HR) = 0.52-0.65), sorafenib (HR = 0.65, 95% confidence interval (CI): 0.45-0.93) or adavosertib (HR = 0.56, 95%CI: 0.35-0.90) significantly improved OS and PFS versus chemotherapy alone. Notably, adavosertib + gemcitabine was associated with an increased risk of grade 3-4 AEs (relative risk (RR) = 1.8, 95%CI: 1.3-2.7), but these were generally manageable. CONCLUSIONS: Bevacizumab-based combinations demonstrate consistent benefits across multiple regimens for PROC. Paclitaxel + bevacizumab emerges as the optimal balance of efficacy and safety. Topotecan + sorafenib could be an alternative for patients who are ineligible for anti-angiogenic therapy.

Humans↗
PubMed2026

Modelling the effects of biological intervention in a dynamical gene network.

Cellular response to environmental and internal signals can be modeled by dynamical gene regulatory networks (GRN). In the literature, three main classes of gene network models can be distinguished: (1) non-quantitative (or data-based) models which do not describe the probability distribution of gene expressions; (2) quantitative models which fully describe the probability distribution of all genes co-expression; and (3) mechanistic models which allow for a causal interpretation of gene interactions. We propose two rigorous frameworks to model gene alteration in a dynamical GRN, depending on whether the network model is quantitative or mechanistic. We explain how these models can be used for design of experiment, or, if additional alteration data are available, for validation purposes or to improve the parameter estimation of the original model. We apply these methods to the Gaussian graphical model, which is quantitative but non-mechanistic, and to mechanistic models of Bayesian networks and penalized linear regression.

Gene Regulatory Networks↗
PubMed2003

Use of gene networks from full genome microarray libraries to identify functionally relevant drug-affected genes and gene regulation cascades.

We developed an extensive yeast gene expression library consisting of full-genome cDNA array data for over 500 yeast strains, each with a single-gene disruption. Using this data, combined with dose and time course expression experiments with the oral antifungal agent griseofulvin, whose exact molecular targets were previously unknown, we used Boolean and Bayesian network discovery techniques to determine the gene expression regulatory cascades affected directly by this drug. Using this method we identified CIK1 as an important affected target gene related to the functional phenotype induced by griseofulvin. Cellular functional analysis of griseofulvin showed similar tubulin-specific morphological effects on mitotic spindle formation to those of the drug, in agreement with the known function of CIK1p. Further, using the nonparametric, nonlinear Bayesian gene networks we were able to identify alternative ligand-dependant transcription factors and G protein homologues upstream of CIK1 that regulate CIK1 expression and might therefore serve as alternative molecular targets to induce the same molecular response as griseofulvin.

Bayes Theorem↗
PubMed2004

Bayesian analysis of signaling networks governing embryonic stem cell fate decisions.

MOTIVATION: Signaling events that direct mouse embryonic stem (ES) cell self-renewal and differentiation are complex and accordingly difficult to understand in an integrated manner. We address this problem by adapting a Bayesian network learning algorithm to model proteomic signaling data for ES cell fate responses to external cues. Using this model we were able to characterize the signaling pathway influences as quantitative, logic-circuit type interactions. Our experimental dataset includes measurements for 28 signaling protein phosphorylation states across 16 different factorial combinations of cytokine and matrix stimuli as reported previously. RESULTS: The Bayesian network modeling approach allows us to uncover previously reported signaling activities related to mouse ES cell self-renewal, such as the roles of LIF and STAT3 in maintaining undifferentiated ES cell populations. Furthermore, the network predicts novel influences such as between ERK phosphorylation and differentiation, or RAF phosphorylation and differentiated cell proliferation. Visualization of the influences detected by the Bayesian network provides intuition about the underlying physiology of the signaling pathways. We demonstrate that the Bayesian networks can capture the linear, nonlinear and multistate logic interactions that connect extracellular cues, intracellular signals and consequent cell functional responses.

Algorithms↗
PubMed1999

Robust QSAR models using Bayesian regularized neural networks.

We describe the use of Bayesian regularized artificial neural networks (BRANNs) in the development of QSAR models. These networks have the potential to solve a number of problems which arise in QSAR modeling such as: choice of model; robustness of model; choice of validation set; size of validation effort; and optimization of network architecture. The application of the methods to QSAR of compounds active at the benzodiazepine and muscarinic receptors is illustrated.

Drug Design↗
PubMed2006

Gene interaction network analysis suggests differences between high and low doses of acetaminophen.

Bayesian networks for quantifying linkages between genes were applied to detect differences in gene expression interaction networks between multiple doses of acetaminophen at multiple time points. Seventeen (17) genes were selected from the gene expression profiles from livers of rats orally exposed to 50, 150 and 1500 mg/kg acetaminophen (APAP) at 6, 24 and 48 h after exposure using a variety of statistical and bioinformatics approaches. The selected genes are related to three biological categories: apoptosis, oxidative stress and other. Gene interaction networks between all 17 genes were identified for the nine dose-time observation points by the TAO-Gen algorithm. Using k-means clustering analysis, the estimated nine networks could be clustered into two consensus networks, the first consisting of the low and middle dose groups, and the second consisting of the high dose. The analysis suggests that the networks could be segregated by doses and were consistent in structure over time of observation within grouped doses. The consensus networks were quantified to calculate the probability distribution for the strength of the linkage between genes connected in the networks. The quantifying analysis showed that, at lower doses, the genes related to the oxidative stress signaling pathway did not interact with the apoptosis-related genes. In contrast, the high-dose network demonstrated significant interactions between the oxidative stress genes and the apoptosis genes and also demonstrated a different network between genes in the oxidative stress pathway. The approaches shown here could provide predictive information to understand high- versus low-dose mechanisms of toxicity.

Acetaminophen↗
PubMed2026

ReGAIN: a bioinformatics platform for assessing probabilistic co-occurrence between resistance genes in bacterial pathogens.

MOTIVATION: Multidrug-resistant bacterial pathogens continue to rise globally, yet scalable methods are needed to infer how resistance determinants co-occur across pathogen populations and to quantify conditional dependencies underlying co-occurrence and shared genetic context. RESULTS: We present ReGAIN (Resistance Gene Association and Inference Network), an open-source platform that applies Bayesian network structure learning to infer probabilistic, conditional dependency relationships among antibiotic resistance, heavy metal tolerance, stress response, and virulence determinants in bacteria. In contrast to pairwise co-occurrence analyses, ReGAIN reports conditional probabilities, relative risks, and absolute risk differences with confidence intervals to prioritize candidate relationships for downstream prioritization. Applied across ESKAPEE pathogens, ReGAIN recapitulated established resistance gene relationships and identified additional candidate patterns consistent with co-selection and shared genetic context. Together, these results support scalable, reproducible population-wide analysis of resistance networks for surveillance, comparative genomics and epidemiology. AVAILABILITY: ReGAIN analyses are performed using Python v3.11.5 and R v4.4.1 and is available as open-source software through Bioconda at {https://anaconda.org/bioconda/regain-cli}. Source code and documentation can be found at {https://github.com/ERBringHorvath/regain_CLI}. All genomes used in this publication were downloaded from the National Center for Biotechnology Information database. Large supplementary tables and results data from the ESKAPEE pathogen example network analyses can be downloaded from https://figshare.com/articles/dataset/ReGAIN_command_line_software_and_supplemental_figures_/28959431.

Computational Biology↗
PubMed2005

Gene network inference from incomplete expression data: transcriptional control of hematopoietic commitment.

MOTIVATION: The topology and function of gene regulation networks are commonly inferred from time series of gene expression levels in cell populations. This strategy is usually invalid if the gene expression in different cells of the population is not synchronous. A promising, though technically more demanding alternative is therefore to measure the gene expression levels in single cells individually. The inference of a gene regulation network requires knowledge of the gene expression levels at successive time points, at least before and after a network transition. However, owing to experimental limitations a complete determination of the precursor state is not possible. RESULTS: We investigate a strategy for the inference of gene regulatory networks from incomplete expression data based on dynamic Bayesian networks. This permits prediction of the number of experiments necessary for network inference depending on parameters including noise in the data, prior knowledge and limited attainability of initial states. Our strategy combines a gradual 'Partial Learning' approach based solely on true experimental observations for the network topology with expectation maximization for the network parameters. We illustrate our strategy by extensive computer simulations in a high-dimensional parameter space in a simulated single-cell-based example of hematopoietic stem cell commitment and in random networks of different sizes. We find that the feasibility of network inferences increases significantly with the experimental ability to force the system into different initial network states, with prior knowledge and with noise reduction. AVAILABILITY: Source code is available under: www.izbi.uni-leipzig.de/services/NetwPartLearn.html SUPPLEMENTARY INFORMATION: Supplementary Data are available at Bioinformatics online.

Algorithms↗
PubMed2004

An empirical Bayes approach to inferring large-scale gene association networks.

MOTIVATION: Genetic networks are often described statistically using graphical models (e.g. Bayesian networks). However, inferring the network structure offers a serious challenge in microarray analysis where the sample size is small compared to the number of considered genes. This renders many standard algorithms for graphical models inapplicable, and inferring genetic networks an 'ill-posed' inverse problem. METHODS: We introduce a novel framework for small-sample inference of graphical models from gene expression data. Specifically, we focus on the so-called graphical Gaussian models (GGMs) that are now frequently used to describe gene association networks and to detect conditionally dependent genes. Our new approach is based on (1) improved (regularized) small-sample point estimates of partial correlation, (2) an exact test of edge inclusion with adaptive estimation of the degree of freedom and (3) a heuristic network search based on false discovery rate multiple testing. Steps (2) and (3) correspond to an empirical Bayes estimate of the network topology. RESULTS: Using computer simulations, we investigate the sensitivity (power) and specificity (true negative rate) of the proposed framework to estimate GGMs from microarray data. This shows that it is possible to recover the true network topology with high accuracy even for small-sample datasets. Subsequently, we analyze gene expression data from a breast cancer tumor study and illustrate our approach by inferring a corresponding large-scale gene association network for 3883 genes.

Algorithms↗
PubMed2007

Analysis of gene networks for drug target discovery and validation.

Understanding responses of the cellular system for a dosing molecule is one of the most important problems in pharmacogenomics. In this chapter, we describe computational methods for identifying and validating drug target genes based on the gene networks estimated from microarray gene expression data. We use two types of microarray gene expression data: gene disruptant microarray data and time-course drug response microarray data. For this purpose, the information of gene networks plays an essential role and is unattainable from clustering methods, which are the standard for gene expression analysis. The gene network is estimated from disruptant microarray data by the Bayesian network model, and then the proposed method automatically identifies sets of genes or gene regulatory pathways affected by the drug. We use an actual example from analysis of Saccharomyces cerevisiae gene expression profile data to express a concrete strategy for the application of gene network information toward drug target discovery.

Antifungal Agents↗
PubMed2005

Identifying drug active pathways from gene networks estimated by gene expression data.

We present a computational method for identifying genes and their regulatory pathways influenced by a drug, using microarray gene expression data collected by single gene disruptions and drug responses. The automatic identification of such genes and pathways in organisms' cells is an important problem for pharmacogenomics and the tailor-made medication. Our method estimates regulatory relationships between genes as a gene network from microarray data of gene disruptions with a Bayesian network model, then identifies the drug affected genes and their regulatory pathways on the estimated network with time course drug response microarray data. Compared to the existing method, our proposed method can identify not only the drug affected genes and the druggable genes, but also the drug responses of the pathways. For evaluating the proposed method, we conducted simulated examples based on artificial networks and expression data. Our method succeeded in identifying the pseudo drug affected genes and pathways with the high coverage greater than 80 %. We also applied our method to Saccharomyces cerevisiae drug response microarray data. In this real example, we identified the genes and the pathways that are potentially influenced by a drug. These computational experiments indicate that our method successfully identifies the drug-activated genes and pathways, and is capable of predicting undesirable side effects of the drug, identifying novel drug target genes, and understanding the unknown mechanisms of the drug.

Bayes Theorem↗
PubMed2004

Finding optimal models for small gene networks.

Finding gene networks from microarray data has been one focus of research in recent years. Given search spaces of super-exponential size, researchers have been applying heuristic approaches like greedy algorithms or simulated annealing to infer such networks. However, the accuracy of heuristics is uncertain, which--in combination with the high measurement noise of microarrays--makes it very difficult to draw conclusions from networks estimated by heuristics. We present a method that finds optimal Bayesian networks of considerable size and show first results of the application to yeast data. Having removed the uncertainty due to the heuristic methods, it becomes possible to evaluate the power of different statistical models to find biologically accurate networks.

Algorithms↗