Representing Hypoexponential Distributions in Continuous Time Bayesian Networks

2018 ◽  
pp. 565-577
Author(s):  
Manxia Liu ◽  
Fabio Stella ◽  
Arjen Hommersom ◽  
Peter J. F. Lucas
Keyword(s):  
Bayesian Networks ◽  
Continuous Time ◽  
Continuous Time Bayesian Networks

scholarly journals Modeling Approaches Reveal New Regulatory Networks in Aspergillus fumigatus Metabolism

2020 ◽  
Vol 6 (3) ◽  
pp. 108
Author(s):  
Enzo Acerbi ◽  
Marcela Hortova-Kohoutkova ◽  
Tsokyi Choera ◽  
Nancy Keller ◽  
Jan Fric ◽  
...  
Keyword(s):  
Bayesian Networks ◽  
Continuous Time ◽  
Regulatory Networks ◽  
Time Course ◽  
Metabolic Networks ◽  
De Novo ◽  
Tryptophan Metabolism ◽  
Reconstruction Method ◽  
Fungal Virulence ◽  
Continuous Time Bayesian Networks

Systems biology approaches are extensively used to model and reverse-engineer gene regulatory networks from experimental data. Indoleamine 2,3-dioxygenases (IDOs)—belonging in the heme dioxygenase family—degrade l-tryptophan to kynurenines. These enzymes are also responsible for the de novo synthesis of nicotinamide adenine dinucleotide (NAD+). As such, they are expressed by a variety of species, including fungi. Interestingly, Aspergillus may degrade l-tryptophan not only via IDO but also via alternative pathways. Deciphering the molecular interactions regulating tryptophan metabolism is particularly critical for novel drug target discovery designed to control pathogen determinants in invasive infections. Using continuous time Bayesian networks over a time-course gene expression dataset, we inferred the global regulatory network controlling l-tryptophan metabolism. The method unravels a possible novel approach to target fungal virulence factors during infection. Furthermore, this study represents the first application of continuous-time Bayesian networks as a gene network reconstruction method in Aspergillus metabolism. The experiment showed that the applied computational approach may improve the understanding of metabolic networks over traditional pathways.

scholarly journals Intrusion Detection using Continuous Time Bayesian Networks

2010 ◽  
Vol 39 ◽  
pp. 745-774 ◽  
Author(s):  
J. Xu ◽  
C. R. Shelton
Keyword(s):  
Intrusion Detection ◽  
Bayesian Networks ◽  
Continuous Time ◽  
Particle Filtering ◽  
Generative Models ◽  
Training Data ◽  
Continuous Time Model ◽  
Time Model ◽  
General Technique ◽  
Continuous Time Bayesian Networks

Intrusion detection systems (IDSs) fall into two high-level categories: network-based systems (NIDS) that monitor network behaviors, and host-based systems (HIDS) that monitor system calls. In this work, we present a general technique for both systems. We use anomaly detection, which identifies patterns not conforming to a historic norm. In both types of systems, the rates of change vary dramatically over time (due to burstiness) and over components (due to service difference). To efficiently model such systems, we use continuous time Bayesian networks (CTBNs) and avoid specifying a fixed update interval common to discrete-time models. We build generative models from the normal training data, and abnormal behaviors are flagged based on their likelihood under this norm. For NIDS, we construct a hierarchical CTBN model for the network packet traces and use Rao-Blackwellized particle filtering to learn the parameters. We illustrate the power of our method through experiments on detecting real worms and identifying hosts on two publicly available network traces, the MAWI dataset and the LBNL dataset. For HIDS, we develop a novel learning method to deal with the finite resolution of system log file time stamps, without losing the benefits of our continuous time model. We demonstrate the method by detecting intrusions in the DARPA 1998 BSM dataset.

scholarly journals Learning Continuous Time Bayesian Networks in Non-stationary Domains

2016 ◽  
Vol 57 ◽  
pp. 1-37 ◽  
Author(s):  
Simone Villa ◽  
Fabio Stella
Keyword(s):  
Bayesian Networks ◽  
Continuous Time ◽  
Learning Algorithm ◽  
State Of The Art ◽  
Synthetic Data ◽  
Score Function ◽  
Dynamic Bayesian Networks ◽  
Continuous Time Bayesian Networks ◽  
Real World Datasets ◽  
Transition Times

Non-stationary continuous time Bayesian networks are introduced. They allow the parents set of each node to change over continuous time. Three settings are developed for learning non-stationary continuous time Bayesian networks from data: known transition times, known number of epochs and unknown number of epochs. A score function for each setting is derived and the corresponding learning algorithm is developed. A set of numerical experiments on synthetic data is used to compare the effectiveness of non-stationary continuous time Bayesian networks to that of non-stationary dynamic Bayesian networks. Furthermore, the performance achieved by non-stationary continuous time Bayesian networks is compared to that achieved by state-of-the-art algorithms on four real-world datasets, namely drosophila, saccharomyces cerevisiae, songbird and macroeconomics.

Sign in / Sign up

Export Citation Format

Share Document