Towards collaborative, intermodal hub networks

2005 ◽  
Vol 41 (6) ◽  
pp. 567-583 ◽  
Author(s):  
Bas Groothedde ◽  
Cees Ruijgrok ◽  
Lóri Tavasszy
Keyword(s):  
Hub Networks

Exact and heuristic algorithms for the design of hub networks with multiple lines

2015 ◽  
Vol 246 (1) ◽  
pp. 186-198 ◽  
Author(s):  
Elisangela Martins de Sá ◽  
Ivan Contreras ◽  
Jean-François Cordeau
Keyword(s):  
Heuristic Algorithms ◽  
Hub Networks

scholarly journals THE EFFECT OF NETWORK TYPES ON HERD BEHAVIOR

2019 ◽  
Vol 4 (8) ◽  
pp. 161-169
Author(s):  
Quji Bichia
Keyword(s):  
Preferential Attachment ◽  
Small World ◽  
Group Behavior ◽  
Human Interaction ◽  
Separate Analysis ◽  
Central Figure ◽  
Two Component ◽  
Hub Networks ◽  
Different Types ◽  
Hub Network

This paper aims to compare spread of an opinion, norm, innovation or a belief in different types of networks. For this purpose, different network metrics are discussed and results of network model are summarized based on simulations. Norms may spread from a single source or multiple sources and these issues require separate analysis. Networks play an important role in decisions that people make. They determine what information someone will receive and how will he act within this limited information. As it turns out, small number of people can influence decisions of majority. These can be consumption decision, decisions about adopting new technologies, innovations, medical practice, social norms and so on. Mathematical models of net- works help us understand how these processes propagate. There are different types of networks that can emerge within a society or some group and there are characteristics that can describe roles of group members in spreading some idea or innovation. Networks can be of many kinds but human networks tend to have common characteristics. Therefore, current work focuses on 4 types of networks - small world, single-hub (one central figure), multi-hub (many central figures) and two-component. Small world random networks are observed in different situations and they can be used to describe some human interaction networks. Many networks are described by power law distributions, where new members of a net- work have a preferential attachment and link to other highly connected members. Single-hub and multi-hub networks describe such situations. Two-component network is used to describe polarized groups that have opposing views and are competing with each other. This could be political parties or competing firms. The present paper analyzes patterns of information flow across different types of networks and compares the conditions for the emergence of group behavior. Contribution of this work is the simulation results that show how different networks exhibit varying outcomes and propagate opinions differently. Simulations on small world, single-hub, multi-hub and two-component networks with 150 members show that net- work types matter in terms of how fast can group behavior spread within a network. The process of spreading group behavior is as follows: Every individual receives some signal si about a binary decision. Individuals make the first decision based on their signals because they have no other information. In the next step, every individual looks at the decisions of those in his or her neighborhood and updates his or her belief by the Bayes rule. On the next step they observe others’ actions again and decide whether to change own action or not and so on. After some stages, a stable point is reached where no one is willing to change his decision anymore. The study compares the times needed to reach stability in different types of networks. Simulations have shown that the speed of propagation of a belief varies according to who is the source of this process. However, the difference is not big within a small world network. As it turns out, full distribution occurs in at least 4 and a maximum of 20 periods, and the average time of full distribution varies from 6.5 to 8.6, depending on whether the most connected member is the source or the least connected one. The result is quite different if there is one central figure. The presence of one central figure prevents information from spreading across the network, as there is preferential attach- ment and some members can only acquire one connection. If there are several central figures, the full spread occurs relatively faster. In a two-component network, full adoption oc- curs quite rapidly. Although the connection between components is almost non-existent, a small number of existing links play a critical role in rapidly disseminating a behavior. Group behavior spreads more rapidly in a random net- work than in a network characteristic of a special society on average. But multi-hub network has the potential for the fast- est spread (although information disseminates faster in a ran- dom network on average). Group behavior is slow to spread in a single-hub network, as some individuals are very weakly connected to other areas of the network. An opinion spread in the neighbourhood of the central figure will soon reach all members of around him or her but it will take a long time to reach far ends of the network. The two-component network in this regard maintains a balance between the speed of dis- tribution and the area of distribution. There is least variation between adoption times in a two-component network (not considering the small-world random network). The high variation in single-hub and multi-hub networks indicates that it is advisable to consider more specific situations for accurate results. Comparison of adoption times within multi-hub net- works of different size shows that adoption happens at the same speed most of the time regardless of the network size. When two opposing opinions are being spread and one of the opinions is dominated by the other, it takes similar time periods for all sizes of multi-hub networks.

Characteristics on hub networks of urban rail transit networks

2016 ◽  
Vol 447 ◽  
pp. 502-507 ◽  
Author(s):  
Jianhua Zhang ◽  
Shuliang Wang ◽  
Zhaojun Zhang ◽  
Kuansheng Zou ◽  
Zhan Shu
Keyword(s):  
Urban Rail Transit ◽  
Rail Transit ◽  
Transit Networks ◽  
Urban Rail ◽  
Hub Networks

Cold chain management in hierarchical operational hub networks

2021 ◽  
Vol 147 ◽  
pp. 102202
Author(s):  
Yalda Esmizadeh ◽  
Mahdi Bashiri ◽  
Hamed Jahani ◽  
Bernardo Almada-Lobo
Keyword(s):  
Cold Chain ◽  
Chain Management ◽  
Hub Networks

scholarly journals Comprehensive Analysis of N6-Methyladenosine RNA Methylation Regulators Expression Identify Distinct Molecular Subtypes of Myocardial Infarction

2021 ◽  
Vol 9 ◽  
Author(s):  
Xin Shi ◽  
Yaochen Cao ◽  
Xiaobin Zhang ◽  
Chang Gu ◽  
Feng Liang ◽  
...  
Keyword(s):  
Myocardial Infarction ◽  
Association Study ◽  
Pathway Analysis ◽  
Genome Wide Association Study ◽  
Molecular Subtypes ◽  
Study Data ◽  
Genome Wide Association ◽  
Single Cell Sequencing ◽  
Genome Wide ◽  
Hub Networks

Background: Myocardial infarction (MI) is one of the leading threats to human health. N6-methyladenosine (m6A) modification, as a pivotal regulator of messenger RNA stability, protein expression, and cellular processes, exhibits important roles in the development of cardiac remodeling and cardiomyocyte contractile function.Methods: The expression levels of m6A regulators were analyzed using the GSE5406 database. We analyzed genome-wide association study data and single-cell sequencing data to confirm the functional importance of m6A regulators in MI. Three molecular subtypes with different clinical characteristics were established to tailor treatment strategies for patients with MI. We applied pathway analysis and differentially expressed gene (DEG) analysis to study the changes in gene expression and identified four common DEGs. Furthermore, we constructed the protein–protein interaction network and confirmed several hub genes in three clusters of MI. To lucubrate the potential functions, we performed a ClueGO analysis of these hub networks.Results: In this study, we identified that the levels of FTO, YTHDF3, ZC3H13, and WTAP were dramatically differently expressed in MI tissues compared with controls. Bioinformatics analysis showed that DEGs in MI were significantly related to modulating calcium signaling and chemokine signaling, and m6A regulators were related to regulating glucose measurement and elevated blood glucose levels. Furthermore, genome-wide association study data analysis showed that WTAP single-nucleotide polymorphism was significantly related to the progression of MI. In addition, single-cell sequencing found that WTAP is widely expressed in the heart tissues. Moreover, we conducted consensus clustering for MI in view of the dysregulated m6A regulators’ expression in MI. According to the expression levels, we found MI patients could be clustered into three subtypes. Pathway analysis showed the DEGs among different clusters in MI were assigned to HIF-1, IL-17, MAPK, PI3K-Akt signaling pathways, etc. The module analysis detected several genes, including BAG2, BAG3, MMP2, etc. We also found that MI-related network was significantly related to positive and negative regulation of angiogenesis and response to heat. The hub networks in MI clusters were significantly related to antigen processing and ubiquitin-mediated proteolysis, RNA splicing, and stability, indicating that these processes may contribute to the development of MI.Conclusion: Collectively, our study could provide more information for understanding the roles of m6A in MI, which may provide a novel insight into identifying biomarkers for MI treatment and diagnosis.

Sign in / Sign up

Export Citation Format

Share Document