scholarly journals Automated Visualization of Rule-based Models

2016 ◽  
Author(s):  
John A. P. Sekar ◽  
Jose-Juan Tapia ◽  
James R. Faeder
Keyword(s):  
Network Architecture ◽  
Molecular Structures ◽  
Signal Pathways ◽  
Graph Visualization ◽  
Chemical Interactions ◽  
Rule Based ◽  
Automated Generation ◽  
Regulatory Interactions ◽  
Rule Set ◽  
Network Compression

AbstractRule-based modeling frameworks provide a specification format in which kinetic interactions are modeled as “reaction rules”. These rules are specified on phosphorylation motifs, domains, binding sites and other sub-molecular structures, and have proved useful for modeling signal transduction. Visual representations are necessary to understand individual rules as well as analyze interactions of hundreds of rules, which motivates the need for automated diagramming tools for rule-based models. Here, we present a theoretical framework that unifies the layers of information in a rule-based model and enables automated visualization of (i) the mechanism encoded in a rule, (ii) the regulatory interaction of two or more rules, and (iii) the emergent network architecture of a large rule set. Specifically, we present a compact rule visualization that conveys the action of a rule explicitly (unlike conventional visualizations), a regulatory graph visualization that conveys regulatory interactions between rules, and a set of graph compression methods that synthesize informative pathway diagrams from complex regulatory graphs. These methods enable inference of network motifs (such as feedback and feed-forward loops), automated generation of signal flow diagrams for hundreds of rules, and tunable network compression using heuristics and graph analysis, all of which are advances over the state of the art for rule-based models. These methods also produce more readable diagrams than currently available tools as we show with an empirical comparison across 27 published rule-based models of various sizes. We provide an implementation in the open source and freely available BioNetGen framework, but the underlying methods are applicable to all current rule-based models in BioNetGen, Kappa and Simmune frameworks. We expect that these tools will promote communication and analysis of rule-based models and their eventual integration into whole cell models.Author SummarySignaling in living cells is mediated through a complex network of chemical interactions. Current predictive models of signal pathways have hundreds of reaction rules that specify chemical interactions, and a comprehensive model of a stem cell or cancer cell would be expected to have many more. Visualizations of rules and their interactions are needed to navigate, organize, communicate and analyze large signaling models. In this work, we have developed: (i) a novel visualization for individual rules that compactly conveys what each rule does, (ii) a comprehensive visualization of a set of rules as a network of regulatory interactions called a regulatory graph, and (iii) a set of procedures for compressing the regulatory graph into a pathway diagram that highlights underlying signaling motifs such as feedback and feedforward loops. We show that these visualizations are compact and informative across models of widely varying sizes. The methods developed here not only improve the understandability of current models, but also establish principles for organizing the much larger models of the future.

scholarly journals Automating Terrain Texturing in Real-Time Using a Rule-Based Approach

2010 ◽  
Vol 9 (4) ◽  
pp. 21-28
Author(s):  
John Ferraris ◽  
Christos Gatzidis ◽  
Feng Tian
Keyword(s):  
Real Time ◽  
Rule Based ◽  
Novel Approach ◽  
Great Performance ◽  
Rule Set ◽  
Complex Features ◽  
Rule Based Approach

This publication proposes a novel approach to automatically colour and texture a given terrain mesh in real time. Through the use of weighting rules, a simple syntax allows for the generation of texture and colour values based on the elevation and angle of a given vertex. It is through this combination of elevation and angle that complex features such as ridges, hills and mountains can be described, with the mesh coloured and textured accordingly. The implementation of the approach is done entirely on the GPU using 2D lookup textures, delivering a great performance increase over typical approaches that pass colour and weighting information in the fragment shader. In fact, the rule set is abstracted enough to be used in conjunction with any colouring/texturing approach that uses weighting values to dictate which surfaces are depicted on the mesh

scholarly journals A New Classification Approach Based on Multiple Classification Rules

2014 ◽  
Vol 2014 ◽  
pp. 1-7 ◽  
Author(s):  
Zhongmei Zhou
Keyword(s):  
Classification Rule ◽  
Classification Rules ◽  
Associative Classification ◽  
New Classification ◽  
Class Label ◽  
Rule Based ◽  
Classification Approach ◽  
Minimum Support ◽  
Multiple Classification ◽  
Rule Set

A good classifier can correctly predict new data for which the class label is unknown, so it is important to construct a high accuracy classifier. Hence, classification techniques are much useful in ubiquitous computing. Associative classification achieves higher classification accuracy than some traditional rule-based classification approaches. However, the approach also has two major deficiencies. First, it generates a very large number of association classification rules, especially when the minimum support is set to be low. It is difficult to select a high quality rule set for classification. Second, the accuracy of associative classification depends on the setting of the minimum support and the minimum confidence. In comparison with associative classification, some improved traditional rule-based classification approaches often produce a classification rule set that plays an important role in prediction. Thus, some improved traditional rule-based classification approaches not only achieve better efficiency than associative classification but also get higher accuracy. In this paper, we put forward a new classification approach called CMR (classification based on multiple classification rules). CMR combines the advantages of both associative classification and rule-based classification. Our experimental results show that CMR gets higher accuracy than some traditional rule-based classification methods.

scholarly journals Classification Based on Pruning and Double Covered Rule Sets for the Internet of Things Applications

2014 ◽  
Vol 2014 ◽  
pp. 1-6
Author(s):  
Shasha Li ◽  
Zhongmei Zhou ◽  
Weiping Wang
Keyword(s):  
Internet Of Things ◽  
High Accuracy ◽  
The Internet ◽  
Training Set ◽  
Rule Based ◽  
Useful Knowledge ◽  
Rule Sets ◽  
Rule Set ◽  
The Internet Of Things

The Internet of things (IOT) is a hot issue in recent years. It accumulates large amounts of data by IOT users, which is a great challenge to mining useful knowledge from IOT. Classification is an effective strategy which can predict the need of users in IOT. However, many traditional rule-based classifiers cannot guarantee that all instances can be covered by at least two classification rules. Thus, these algorithms cannot achieve high accuracy in some datasets. In this paper, we propose a new rule-based classification, CDCR-P (Classification based on the Pruning and Double Covered Rule sets). CDCR-P can induce two different rule setsAandB. Every instance in training set can be covered by at least one rule not only in rule setA, but also in rule setB. In order to improve the quality of rule setB, we take measure to prune the length of rules in rule setB. Our experimental results indicate that, CDCR-P not only is feasible, but also it can achieve high accuracy.

Rule-based fuzzy inference system for estimating the influent COD/N ratio and ammonia load to a sequencing batch reactor

2006 ◽  
Vol 53 (1) ◽  
pp. 199-207 ◽  
Author(s):  
Y.J. Kim ◽  
H. Bae ◽  
J.H. Ko ◽  
K.M. Poo ◽  
S. Kim ◽  
...  
Keyword(s):  
Fuzzy Inference System ◽  
Sequencing Batch Reactor ◽  
Fuzzy Inference ◽  
Batch Reactor ◽  
Rule Generation ◽  
Decision Tree Algorithm ◽  
Inference System ◽  
Rule Based ◽  
Close Relationship ◽  
Rule Set

A fuzzy inference system using sensor measurements was developed to estimate the influent COD/N ratio and ammonia load. The sensors measured ORP, DO and pH. The sensor profiles had a close relationship with the influent COD/N ratio and ammonia load. To confirm this operational knowledge for constructing a rule set, a correlation analysis was conducted. The results showed that a rule generation method based only on operational knowledge did not generate a sufficiently accurate relationship between sensor measurements and target variables. To compensate for this defect, a decision tree algorithm was used as a standardized method for rule generation. Given a set of inputs, this algorithm was used to determine the output variables. However, the generated rules could not estimate the continuous influent COD/N ratio and ammonia load. Fuzzified rules and the fuzzy inference system were developed to overcome this problem. The fuzzy inference system estimated the influent COD/N ratio and ammonia load quite well. When these results were compared to the results from a predictive polynomial neural network model, the fuzzy inference system was more stable.

GA Based Hybrid Fuzzy Rule Optimization Approach for Elevator Group Control System

2013 ◽  
Vol 284-287 ◽  
pp. 2380-2384 ◽  
Author(s):  
Ta Cheng Chen ◽  
Yuan Yong Hsu ◽  
An Chen Lee ◽  
Shiang Yu Wang
Keyword(s):  
Control System ◽  
Flow Direction ◽  
Fuzzy Rule ◽  
Fuzzy Rules ◽  
Optimization Approach ◽  
Department Stores ◽  
Rule Based ◽  
Group Control ◽  
Rule Set ◽  
Elevator Group Control

Elevators are the essential transportation tools in high buildings so that Elevator Group Control System (EGCS) is developed to dynamically layout the schedule of elevators in a group. In this study, a fuzzy rules based intelligent elevator group control system has been proposed in which the structure of rules including the related parameters are generated optimally based on the traffic data so as to maximize service quality. In literature, the fuzzy related approaches have been applied in EGCS but the fuzzy rules were all pre-defined. However, how to create the most suitable fuzzy rule set in EGCS for dispatching elevators more efficiently and economically are never discussed in literature. The aim of the proposed approach is to minimize the average waiting time at peak hours as well as to minimize the power energy at off-peak hours by using the proposed fuzzy rule based ECGS. Moreover, there are many decision variables are considered in the GCGS to provide the most appropriate elevator assignment whenever any hall call is given. These variables include the number of elevators, traffic flow, direction, passenger preferences (for instance, department stores, hospitals, hotels, and office buildings), congestion and VIP priority floor, etc. In this study, a fuzzy rule based elevator-dispatching approach has been proposed for the EGCS in which the fuzzy rules and related parameters are derived optimally by using genetic algorithm based on the historical elevator transportation data. The experimental results show that the performance of the proposed approach is superior to these of traditional approaches in literatures.

scholarly journals Rule-Based Neural Networks for Classification and Probability Estimation

1992 ◽  
Vol 4 (6) ◽  
pp. 781-804 ◽  
Author(s):  
Rodney M. Goodman ◽  
Charles M. Higgins ◽  
John W. Miller ◽  
Padhraic Smyth
Keyword(s):  
Neural Network ◽  
Network Architecture ◽  
Domain Knowledge ◽  
Probability Estimation ◽  
Theoretic Approach ◽  
Rule Based ◽  
Probability Estimates ◽  
Neural Network Classifiers ◽  
Information Theoretic Approach ◽  
Rule Based Approach

In this paper we propose a network architecture that combines a rule-based approach with that of the neural network paradigm. Our primary motivation for this is to ensure that the knowledge embodied in the network is explicitly encoded in the form of understandable rules. This enables the network's decision to be understood, and provides an audit trail of how that decision was arrived at. We utilize an information theoretic approach to learning a model of the domain knowledge from examples. This model takes the form of a set of probabilistic conjunctive rules between discrete input evidence variables and output class variables. These rules are then mapped onto the weights and nodes of a feedforward neural network resulting in a directly specified architecture. The network acts as parallel Bayesian classifier, but more importantly, can also output posterior probability estimates of the class variables. Empirical tests on a number of data sets show that the rule-based classifier performs comparably with standard neural network classifiers, while possessing unique advantages in terms of knowledge representation and probability estimation.

Neural Network Model for Predicting Anticancer Activity of Pyridopyrimidines Derivatives

2014 ◽  
Vol 905 ◽  
pp. 96-100 ◽  
Author(s):  
Xi Hhua Du ◽  
Wen Chang Zhuang
Keyword(s):  
Neural Network ◽  
Anticancer Activity ◽  
Network Model ◽  
Neural Network Model ◽  
Network Architecture ◽  
Structural Parameters ◽  
Back Propagation ◽  
Molecular Structures ◽  
Molecular Connectivity ◽  
Molecular Connectivity Index

Molecular structures of pyridopyrimidines derivatives as known as dihydrofolate reductase (DHFR) inhibitors were investigated by using the neural network method. Based on the molecular connectivity, molecular connectivity index and molecular electronegativity distance vectors of 32 pyridopyrimidine derivatives were obtained. Among these parameters, three optimized structural parameters 1χ3χ and M17 as the input neurons of the artificial neural network were selected by step-wise regression. Then a 3:4:1 network architecture was employed and a satisfying neural network model for predicting anticancer activity (lg1/C) was constructed with the back-propagation (BP) algorithm. The total correlation coefficient R and the standard deviation S were 0.925 and 0.336 respectively that showed significantly nonlinear relationships between lg1/C and three structural parameters. It was concluded that the predictions of BP neural network are better than those of methods in the literatures.

Traditional Chinese Medicine: A Treasured Natural Resource of Anticancer Drug Research and Development

2014 ◽  
Vol 42 (03) ◽  
pp. 543-559 ◽  
Author(s):  
Chao-Yun Wang ◽  
Xian-Yong Bai ◽  
Chun-Hua Wang
Keyword(s):  
Chinese Medicine ◽  
Traditional Chinese Medicine ◽  
The United States ◽  
Natural Compounds ◽  
Molecular Structures ◽  
New Drugs ◽  
Signal Pathways ◽  
Active Ingredients ◽  
Antitumor Properties

To discover and develop novel natural compounds, active ingredients, single herbs and combination formulas or prescriptions in traditional Chinese medicine (TCM) with therapeutic selectivity that can preferentially kill cancer cells and inhibit the amplification of cancer without significant toxicity is an important area in cancer therapy. A lot of valuable TCMs were applied as alternative or complementary medicines in the United States and Europe. But these TCMs, as one of the main natural resources, were widely used to research and develop new drugs in Asia. In TCMs, some specific herbs, animals, minerals and combination formulas were recorded and exploited due to their active ingredients and specific natural compounds with antitumor activities. The article focused on the antitumor properties of natural compounds and combination formulas or prescriptions in TCMs, described its influence on tumor progression, angiogenesis, metastasis, and revealed its mechanisms of antitumor and inhibitory action. Among the nature compounds, triptolide, berberine, matrine, oxymatrine, kurarinone and deoxypodophyllotoxin (DPT) with specific molecular structures have been separated, purified, and evaluated their antitumor properties in vitro and in vivo. Cancer is a multifactorial and multistep disease, so the treatment effect of combination formulas and prescriptions in TCMs involving multi-targets and multi-signal pathways on tumor may be superior than that of agents targeting a single molecular target alone. Shi Quan Da Bu Tang and Yanshu injection, as well known combination formulas and prescriptions in TCMs, have shown an excellent therapeutic effect on cancer.

Sign in / Sign up

Export Citation Format

Share Document