Causal Analysis of Learning Performance Based on Bayesian Network and Mutual Information
Ontology highlight
ABSTRACT:
SUBMITTER: Chen J
PROVIDER: S-EPMC7514446 | biostudies-literature | 2019 Nov
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:MotivationFunctional protein-protein interaction (PPI) networks elucidate molecular pathways underlying complex phenotypes, including those of human diseases. Extrapolation of domain-domain interactions (DDIs) from known PPIs is a major domain-based method for inferring functional PPI networks. However, the protein domain is a functional unit of the protein. Therefore, we should be able to effectively infer functional interactions between proteins based on the co-occurrence of domains.ResultsHere, we present a method for inferring accurate functional PPIs based on the similarity of domain composition between proteins by weighted mutual information (MI) that assigned different weights to the domains based on their genome-wide frequencies. Weighted MI outperforms other domain-based network inference methods and is highly predictive for pathways as well as phenotypes. A genome-scale human functional network determined by our method reveals numerous communities that are significantly associated with known pathways and diseases. Domain-based functional networks may, therefore, have potential applications in mapping domain-to-pathway or domain-to-phenotype associations.Availability and implementationSource code for calculating weighted mutual information based on the domain profile matrix is available from www.netbiolab.org/w/WMI CONTACT: Insuklee@yonsei.ac.krSupplementary informationSupplementary data are available at Bioinformatics online.
| S-EPMC5018372 | biostudies-literature
Project description:BackgroundThe problems of correlation and classification are long-standing in the fields of statistics and machine learning, and techniques have been developed to address these problems. We are now in the era of high-dimensional data, which is data that can concern billions of variables. These data present new challenges. In particular, it is difficult to discover predictive variables, when each variable has little marginal effect. An example concerns Genome-wide Association Studies (GWAS) datasets, which involve millions of single nucleotide polymorphism (SNPs), where some of the SNPs interact epistatically to affect disease status. Towards determining these interacting SNPs, researchers developed techniques that addressed this specific problem. However, the problem is more general, and so these techniques are applicable to other problems concerning interactions. A difficulty with many of these techniques is that they do not distinguish whether a learned interaction is actually an interaction or whether it involves several variables with strong marginal effects.Methodology/findingsWe address this problem using information gain and Bayesian network scoring. First, we identify candidate interactions by determining whether together variables provide more information than they do separately. Then we use Bayesian network scoring to see if a candidate interaction really is a likely model. Our strategy is called MBS-IGain. Using 100 simulated datasets and a real GWAS Alzheimer's dataset, we investigated the performance of MBS-IGain.Conclusions/significanceWhen analyzing the simulated datasets, MBS-IGain substantially out-performed nine previous methods at locating interacting predictors, and at identifying interactions exactly. When analyzing the real Alzheimer's dataset, we obtained new results and results that substantiated previous findings. We conclude that MBS-IGain is highly effective at finding interactions in high-dimensional datasets. This result is significant because we have increasingly abundant high-dimensional data in many domains, and to learn causes and perform prediction/classification using these data, we often must first identify interactions.
| S-EPMC4666609 | biostudies-literature
Project description:Motivation:Genome-wide measurements of paired miRNA and gene expression data have enabled the prediction of competing endogenous RNAs (ceRNAs). It has been shown that the sponge effect mediated by protein-coding as well as non-coding ceRNAs can play an important regulatory role in the cell in health and disease. Therefore, many computational methods for the computational identification of ceRNAs have been suggested. In particular, methods based on Conditional Mutual Information (CMI) have shown promising results. However, the currently available implementation is slow and cannot be used to perform computations on a large scale. Results:Here, we present JAMI, a Java tool that uses a non-parametric estimator for CMI values from gene and miRNA expression data. We show that JAMI speeds up the computation of ceRNA networks by a factor of ?70 compared to currently available implementations. Further, JAMI supports multi-threading to make use of common multi-core architectures for further performance gain. Requirements:Java 8. Availability and implementation:JAMI is available as open-source software from https://github.com/SchulzLab/JAMI. Supplementary information:Supplementary data are available at Bioinformatics online.
| S-EPMC6129307 | biostudies-literature
Project description:MotivationStudying the interaction or co-expression of the proteins or markers in the tumor microenvironment of cancer subjects can be crucial in the assessment of risks, such as death or recurrence. In the conventional approach, the cells need to be declared positive or negative for a marker based on its intensity. For multiple markers, manual thresholds are required for all the markers, which can become cumbersome. The performance of the subsequent analysis relies heavily on this step and thus suffers from subjectivity and lacks robustness.ResultsWe present a new method where different marker intensities are viewed as dependent random variables, and the mutual information (MI) between them is considered to be a metric of co-expression. Estimation of the joint density, as required in the traditional form of MI, becomes increasingly challenging as the number of markers increases. We consider an alternative formulation of MI which is conceptually similar but has an efficient estimation technique for which we develop a new generalization. With the proposed method, we analyzed a lung cancer dataset finding the co-expression of the markers, HLA-DR and CK to be associated with survival. We also analyzed a triple negative breast cancer dataset finding the co-expression of the immuno-regulatory proteins, PD1, PD-L1, Lag3 and IDO, to be associated with disease recurrence. We demonstrated the robustness of our method through different simulation studies.Availability and implementationThe associated R package can be found here, https://github.com/sealx017/MIAMI.Supplementary informationSupplementary data are available at Bioinformatics online.
| S-EPMC9344855 | biostudies-literature
Project description:BackgroundNetwork-based approaches have recently gained considerable popularity in high- dimensional regression settings. For example, the Cox regression model is widely used in expression analysis to predict the survival of patients. However, as the number of genes becomes substantially larger than the number of samples, the traditional Cox or L2-regularized Cox models are still prone to noise and produce unreliable estimations of regression coefficients. A recent approach called the network-based Cox (Net-Cox) model attempts to resolve this issue by incorporating prior gene network information into the Cox regression. The Net-Cox model has shown to outperform the models that do not use this network information.ResultsIn this study, we demonstrate an alternative network construction method for the outcome-guided gene interaction network, and we investigate its utility in survival analysis using Net-Cox regression as compared with conventional networks, such as co-expression or static networks obtained from the existing knowledgebase. Our network edges consist of gene pairs that are significantly associated with the clinical outcome. We measure the strength of this association using mutual information between the gene pair and the clinical outcome. We applied this approach to ovarian cancer patients' data in The Cancer Genome Atlas (TCGA) and compared the predictive performance of the proposed approach with those that use other types of networks.ConclusionsWe found that the alternative outcome-guided mutual information network further improved the prediction power of the network-based Cox regression. We expect that a modification of the network regularization term in the Net-Cox model could further improve its prediction power because the properties of our network edges are not optimally reflected in its current form.
| S-EPMC4331683 | biostudies-literature
Project description:Genome-wide association study (GWAS) has turned out to be an essential technology for exploring the genetic mechanism of complex traits. To reduce the complexity of computation, it is well accepted to remove unrelated single nucleotide polymorphisms (SNPs) before GWAS, e.g., by using iterative sure independence screening expectation-maximization Bayesian Lasso (ISIS EM-BLASSO) method. In this work, a modified version of ISIS EM-BLASSO is proposed, which reduces the number of SNPs by a screening methodology based on Pearson correlation and mutual information, then estimates the effects via EM-Bayesian Lasso (EM-BLASSO), and finally detects the true quantitative trait nucleotides (QTNs) through likelihood ratio test. We call our method a two-stage mutual information based Bayesian Lasso (MBLASSO). Under three simulation scenarios, MBLASSO improves the statistical power and retains the higher effect estimation accuracy when comparing with three other algorithms. Moreover, MBLASSO performs best on model fitting, the accuracy of detected associations is the highest, and 21 genes can only be detected by MBLASSO in Arabidopsis thaliana datasets.
| S-EPMC7516787 | biostudies-literature
Project description:A novel information-theoretic approach is proposed to assess the global practical identifiability of Bayesian statistical models. Based on the concept of conditional mutual information, an estimate of information gained for each model parameter is used to quantify the identifiability with practical considerations. No assumptions are made about the structure of the statistical model or the prior distribution while constructing the estimator. The estimator has the following notable advantages: first, no controlled experiment or data is required to conduct the practical identifiability analysis; second, unlike popular variance-based global sensitivity analysis methods, different forms of uncertainties, such as model-form, parameter, or measurement can be taken into account; third, the identifiability analysis is global, and therefore independent of a realization of the parameters. If an individual parameter has low identifiability, it can belong to an identifiable subset such that parameters within the subset have a functional relationship and thus have a combined effect on the statistical model. The practical identifiability framework is extended to highlight the dependencies between parameter pairs that emerge a posteriori to find identifiable parameter subsets. The applicability of the proposed approach is demonstrated using a linear Gaussian model and a non-linear methane-air reduced kinetics model. It is shown that by examining the information gained for each model parameter along with its dependencies with other parameters, a subset of parameters that can be estimated with high posterior certainty can be found.
| S-EPMC10603099 | biostudies-literature
Project description:We propose and demonstrate a representation learning approach by maximizing the mutual information between local features of images and text. The goal of this approach is to learn useful image representations by taking advantage of the rich information contained in the free text that describes the findings in the image. Our method trains image and text encoders by encouraging the resulting representations to exhibit high local mutual information. We make use of recent advances in mutual information estimation with neural network discriminators. We argue that the sum of local mutual information is typically a lower bound on the global mutual information. Our experimental results in the downstream image classification tasks demonstrate the advantages of using local features for image-text representation learning. Our code is available at: https://github.com/RayRuizhiLiao/mutual_info_img_txt.
| S-EPMC9576150 | biostudies-literature
Project description:BackgroundFinding correlation patterns is an important goal of analyzing biological data. Currently available methods for correlation analysis mainly use non-direct associations, such as the Pearson correlation coefficient, and focus on the interpretation of networks at the level of modules. For biological objects such as genes, their collective function depends on pairwise gene-to-gene interactions. However, a large amount of redundant results from module level methods often necessitate further detailed analysis of gene interactions. New approaches of measuring direct associations among variables, such as the part mutual information (PMI), may help us better interpret the correlation pattern of biological data at the level of variable pairs.ResultsWe use PMI to calculate gene co-expression networks of cancer mRNA transcriptome data. Our results show that the PMI-based networks with fewer edges could represent the correlation pattern and are robust across biological conditions. The PMI-based networks recall significantly more important parts of omics defined gene-pair relationships than the Pearson Correlation Coefficient (PCC)-based networks. Based on the scores derived from PMI-recalled copy number variation or DNA methylation gene-pairs, the patients with cancer can be divided into groups with significant differences on disease specific survival.ConclusionsPMI, measuring direct associations between variables, extracts more important biological relationships at the level of gene pairs than conventional indirect association measures do. It can be used to refine module level results from other correlation methods. Particularly, PMI is beneficial to analysis of biological data of the complicated systems, for example, cancer transcriptome data.
| S-EPMC9128248 | biostudies-literature
Project description:The accurate classification of microbes is critical in today's context for monitoring the ecological balance of a habitat. Hence, in this research work, a novel method to automate the process of identifying microorganisms has been implemented. To extract the bodies of microorganisms accurately, a generalized segmentation mechanism which consists of a combination of convolution filter (Kirsch) and a variance-based pixel clustering algorithm (Otsu) is proposed. With exhaustive corroboration, a set of twenty-five features were identified to map the characteristics and morphology for all kinds of microbes. Multiple techniques for feature selection were tested and it was found that mutual information (MI)-based models gave the best performance. Exhaustive hyperparameter tuning of multilayer layer perceptron (MLP), k-nearest neighbors (KNN), quadratic discriminant analysis (QDA), logistic regression (LR), and support vector machine (SVM) was done. It was found that SVM radial required further improvisation to attain a maximum possible level of accuracy. Comparative analysis between SVM and improvised SVM (ISVM) through a 10-fold cross validation method ultimately showed that ISVM resulted in a 2% higher performance in terms of accuracy (98.2%), precision (98.2%), recall (98.1%), and F1 score (98.1%).
| S-EPMC7927045 | biostudies-literature