A comparative analysis of link removal strategies in real complex weighted networks

Land Use Change Disrupts the Network Complexity and Stability of Soil Microbial Carbon Cycling Genes Across an Agricultural Mosaic Landscape

To understand the effects of agricultural land use change and management on soil carbon (C) cycling, it is crucial to examine how these changes can influence microbial soil C cycling. Network analysis can offer insights into the structure, complexity, and stability of the soil microbiome in response to environmental disturbances, including land use change. Using SparCC-based co-occurrence networks, we studied how land use change impacts the connectivity, complexity, and stability of microbial C-cycling gene networks across an agricultural mosaic landscape in Canterbury, New Zealand. The most densely connected networks were found in land uses that were under the most intensive agricultural management, or under naturally regenerating vegetation. The microbial C-cycling gene networks from both land uses presented high network connectivity, low modularity, and a low proportion of negative gene interactions. In contrast, microbial C-cycling genes from native forests, which had the most stable and undisturbed plant cover, had the lowest network connectivity, highest modularity, and a greater proportion of negative gene interactions. Although the differences in total soil C content between land uses were small, the large effects of land use on the network structure of microbial C-cycling genes may have important implications for long-term microbial soil C cycling. Furthermore, this research highlights the value of using microbial network analysis to study the metabolic gene interactions shaping the functional structure of soil microbial communities in a manner not typically captured by more traditional forms of microbial diversity analysis.

From Homeostasis to Neuroinflammation: Insights into Cellular and Molecular Interactions and Network Dynamics

Neuroinflammation is a complex and multifaceted process that involves dynamic interactions among various cellular and molecular components. This sophisticated interplay supports both environmental adaptability and system resilience in the central nervous system (CNS) but may be disrupted during neuroinflammation. In this article, we first characterize the key players in neuroimmune interactions, including microglia, astrocytes, neurons, immune cells, and essential signaling molecules such as cytokines, neurotransmitters, extracellular matrix (ECM) components, and neurotrophic factors. Under homeostatic conditions, these elements promote cellular cooperation and stability, whereas in neuroinflammatory states, they drive adaptive responses that may become pathological if dysregulated. We examine how neuroimmune interactions, mediated through these cellular actors and signaling pathways, create complex networks that regulate CNS functionality and respond to injury or inflammation. To further elucidate these dynamics, we provide insights using a multilayer network (MLN) approach, highlighting the interconnected nature of neuroimmune interactions under both inflammatory and homeostatic conditions. This perspective aims to enhance our understanding of neuroimmune communication and the mechanisms underlying shifts from homeostasis to neuroinflammation. Applying an MLN approach offers a more integrative view of CNS resilience and adaptability, helping to clarify inflammatory processes and identify novel intervention points within the layered landscape of neuroinflammatory responses.

Machine learning in obsessive-compulsive disorder medications

Obsessive-compulsive disorder (OCD) is the fourth most common psychiatric disorder with a significant morbidity rate. Despite various treatment modalities and medications, some patients show no definitive response. The aim of this study is to classify the medications of OCD with machine learning (ML) methods and to compare the classification performances of the decision tree (DT), chi-square automatic interaction detection (CHAID) algorithm, and linear model in ML methods. This research is a descriptive analytical study based on co-word and artificial intelligence methods. The DT models were created with a target (total weight link strength). For hyperparameter optimization, the Gini index was used as the weight total link strength. The performance of the DT model was evaluated based on the prediction model. A total of 116 drugs were extracted from 6574 articles based on co-word analysis, and 56 drugs were classified as the DT's root. These drugs were categorized into six groups in the EWKM diagram. The DT was constructed using the weight.total.link index, with 7 items in Label 3 and 42 items in Label 5 serving as DT leaves. The ML analysis provided valuable insights into the efficacy of various medications such as clomipramine, duloxetine, and pindolol, as well as supplements such as folate, in the treatment of OCD. Treating concomitant diseases, namely hypothyroidism and streptococcal infection could improve the efficacy of treatment.

Hyper-brain hyper-frequency network topology dynamics when playing guitar in quartet

Ensemble music performance is a highly coordinated form of social behavior requiring not only precise motor actions but also synchronization of different neural processes both within and between the brains of ensemble players. In previous analyses, which were restricted to within-frequency coupling (WFC), we showed that different frequencies participate in intra- and inter-brain coordination, exhibiting distinct network topology dynamics that underlie coordinated actions and interactions. However, many of the couplings both within and between brains are likely to operate across frequencies. Hence, to obtain a more complete picture of hyper-brain interaction when musicians play the guitar in a quartet, cross-frequency coupling (CFC) has to be considered as well. Furthermore, WFC and CFC can be used to construct hyper-brain hyper-frequency networks (HB-HFNs) integrating all the information flows between different oscillation frequencies, providing important details about ensemble interaction in terms of network topology dynamics (NTD). Here, we reanalyzed EEG (electroencephalogram) data obtained from four guitarists playing together in quartet to explore changes in HB-HFN topology dynamics and their relation to acoustic signals of the music. Our findings demonstrate that low-frequency oscillations (e.g., delta, theta, and alpha) play an integrative or pacemaker role in such complex networks and that HFN topology dynamics are specifically related to the guitar quartet playing dynamics assessed by sound properties. Simulations by link removal showed that the HB-HFN is relatively robust against loss of connections, especially when the strongest connections are preserved and when the loss of connections only affects the brain of one guitarist. We conclude that HB-HFNs capture neural mechanisms that support interpersonally coordinated action and behavioral synchrony.

Community cohesion looseness in gene networks reveals individualized drug targets and resistance

Community cohesion plays a critical role in the determination of an individual's health in social science. Intriguingly, a community structure of gene networks indicates that the concept of community cohesion could be applied between the genes as well to overcome the limitations of single gene-based biomarkers for precision oncology. Here, we develop community cohesion scores which precisely quantify the community ability to retain the interactions between the genes and their cellular functions in each individualized gene network. Using breast cancer as a proof-of-concept study, we measure the community cohesion score profiles of 950 case samples and predict the individualized therapeutic targets in 2-fold. First, we prioritize them by finding druggable genes present in the community with the most and relatively decreased scores in each individual. Then, we pinpoint more individualized therapeutic targets by discovering the genes which greatly contribute to the community cohesion looseness in each individualized gene network. Compared with the previous approaches, the community cohesion scores show at least four times higher performance in predicting effective individualized chemotherapy targets based on drug sensitivity data. Furthermore, the community cohesion scores successfully discover the known breast cancer subtypes and we suggest new targeted therapy targets for triple negative breast cancer (e.g. KIT and GABRP). Lastly, we demonstrate that the community cohesion scores can predict tamoxifen responses in ER+ breast cancer and suggest potential combination therapies (e.g. NAMPT and RXRA inhibitors) to reduce endocrine therapy resistance based on individualized characteristics. Our method opens new perspectives for the biomarker development in precision oncology.

The time-evolving epileptic brain network: concepts, definitions, accomplishments, perspectives

Epilepsy is now considered a network disease that affects the brain across multiple levels of spatial and temporal scales. The paradigm shift from an epileptic focus-a discrete cortical area from which seizures originate-to a widespread epileptic network-spanning lobes and hemispheres-considerably advanced our understanding of epilepsy and continues to influence both research and clinical treatment of this multi-faceted high-impact neurological disorder. The epileptic network, however, is not static but evolves in time which requires novel approaches for an in-depth characterization. In this review, we discuss conceptual basics of network theory and critically examine state-of-the-art recording techniques and analysis tools used to assess and characterize a time-evolving human epileptic brain network. We give an account on current shortcomings and highlight potential developments towards an improved clinical management of epilepsy.

An optimization-based algorithm for obtaining an optimal synchronizable network after link addition or reduction

Achieving a network structure with optimal synchronization is essential in many applications. This paper proposes an optimization algorithm for constructing a network with optimal synchronization. The introduced algorithm is based on the eigenvalues of the connectivity matrix. The performance of the proposed algorithm is compared with random link addition and a method based on the eigenvector centrality. It is shown that the proposed algorithm has a better synchronization ability than the other methods and also the scale-free and small-world networks with the same number of nodes and links. The proposed algorithm can also be applied for link reduction while less disturbing its synchronization. The effectiveness of the algorithm is compared with four other link reduction methods. The results represent that the proposed algorithm is the most appropriate method for preserving synchronization.

Uncertainty in vulnerability of networks under attack

This study builds conceptual explanations and empirical examinations of the vulnerability response of networks under attack. Two quantities of "vulnerability" and "uncertainty in vulnerability" are defined by scrutinizing the performance loss trajectory of networks experiencing attacks. Both vulnerability and uncertainty in vulnerability quantities are a function of the network topology and size. This is tested on 16 distinct topologies appearing in infrastructure, social, and biological networks with 8 to 26 nodes under two percolation scenarios exemplifying benign and malicious attacks. The findings imply (i) crossing path, tree, and diverging tail are the most vulnerable topologies, (ii) complete and matching pairs are the least vulnerable topologies, (iii) complete grid and complete topologies show the most uncertainty for vulnerability, and (iv) hub-and-spoke and double u exhibit the least uncertainty in vulnerability. The findings also imply that both vulnerability and uncertainty in vulnerability increase with an increase in the size of the network. It is argued that in networks with no undirected cycle and one undirected cycle, the uncertainty in vulnerability is maximal earlier in the percolation process. With an increase in the number of cycles, the uncertainty in vulnerability is accumulated at the end of the percolation process. This emphasizes the role of tailoring preparedness, response, and recovery phases for networks with different topologies when they might experience disruption.

Contact networks have small metric backbones that maintain community structure and are primary transmission subgraphs

The structure of social networks strongly affects how different phenomena spread in human society, from the transmission of information to the propagation of contagious diseases. It is well-known that heterogeneous connectivity strongly favors spread, but a precise characterization of the redundancy present in social networks and its effect on the robustness of transmission is still lacking. This gap is addressed by the metric backbone, a weight- and connectivity-preserving subgraph that is sufficient to compute all shortest paths of weighted graphs. This subgraph is obtained via algebraically-principled axioms and does not require statistical sampling based on null-models. We show that the metric backbones of nine contact networks obtained from proximity sensors in a variety of social contexts are generally very small, 49% of the original graph for one and ranging from about 6% to 20% for the others. This reflects a surprising amount of redundancy and reveals that shortest paths on these networks are very robust to random attacks and failures. We also show that the metric backbone preserves the full distribution of shortest paths of the original contact networks-which must include the shortest inter- and intra-community distances that define any community structure-and is a primary subgraph for epidemic transmission based on pure diffusion processes. This suggests that the organization of social contact networks is based on large amounts of shortest-path redundancy which shapes epidemic spread in human populations. Thus, the metric backbone is an important subgraph with regard to epidemic spread, the robustness of social networks, and any communication dynamics that depend on complex network shortest paths.

A comparison of node vaccination strategies to halt SIR epidemic spreading in real-world complex networks

We compared seven node vaccination strategies in twelve real-world complex networks. The node vaccination strategies are modeled as node removal on networks. We performed node vaccination strategies both removing nodes according to the initial network structure, i.e., non-adaptive approach, and performing partial node rank recalculation after node removal, i.e., semi-adaptive approach. To quantify the efficacy of each vaccination strategy, we used three epidemic spread indicators: the size of the largest connected component, the total number of infected at the end of the epidemic, and the maximum number of simultaneously infected individuals. We show that the best vaccination strategies in the non-adaptive and semi-adaptive approaches are different and that the best strategy also depends on the number of available vaccines. Furthermore, a partial recalculation of the node centrality increases the efficacy of the vaccination strategies by up to 80%.

Reversibility of link prediction and its application to epidemic mitigation

Current link prediction strategies are about finding new probable strong relations to establish or weak ones to remove. An interesting strategy is utilizing link prediction to prioritize the edges in the network and finding newly probable established relations. In this paper we will introduce and explain RLP, reverse link prediction, as a new paradigm, and use popular basic scoring methods including CN, JC, AA, RA, and PA, as its core to examine. The test cases are nine datasets. Half of them are contact networks in different levels from personal contact to aviation, and another half is for covering different test situations. After reviewing the edge removal based epidemic mitigation methods, we show that RLP can be used to decrease the epidemics spreading speed as a general method with various link prediction algorithms, and here in this paper, preferential attachment (PA) has the best results overall. But the results heavily depend on the nature of the examined networks: regular, scale-free or small-world. We also propose an easy to understand criteria, path count, for comparing the efficacy of epidemics mitigation methods. RLP can be extended to use other link prediction scoring methods in various types of graphs as well.

Network Theory and Switching Behaviors: A User Guide for Analyzing Electronic Records Databases

As part of studies that employ health electronic records databases, this paper advocates the employment of graph theory for investigating drug-switching behaviors. Unlike the shared approach in this field (comparing groups that have switched with control groups), network theory can provide information about actual switching behavior patterns. After a brief and simple introduction to fundamental concepts of network theory, here we present (i) a Python script to obtain an adjacency matrix from a records database and (ii) an illustrative example of the application of network theory basic concepts to investigate drug-switching behaviors. Further potentialities of network theory (weighted matrices and the use of clustering algorithms), along with the generalization of these methods to other kinds of switching behaviors beyond drug switching, are discussed.

Human mobility in response to COVID-19 in France, Italy and UK

The COVID-19 pandemic is one of the defining events of our time. National Governments responded to the global crisis by implementing mobility restrictions to slow down the spread of the virus. To assess the impact of those policies on human mobility, we perform a massive comparative analysis on geolocalized data from 13 M Facebook users in France, Italy, and the UK. We find that lockdown generally affects national mobility efficiency and smallworldness-i.e., a substantial reduction of long-range connections in favor of local paths. The impact, however, differs among nations according to their mobility infrastructure. We find that mobility is more concentrated in France and UK and more distributed in Italy. In this paper we provide a framework to quantify the substantial impact of the mobility restrictions. We introduce a percolation model mimicking mobility network disruption and find that node persistence in the percolation process is significantly correlated with the economic and demographic characteristics of countries: areas showing higher resilience to mobility disruptions are those where Value Added per Capita and Population Density are high. Our methods and findings provide important insights to enhance preparedness for global critical events and to incorporate resilience as a relevant dimension to estimate the socio-economic consequences of mobility restriction policies.

Modelling photosystem I as a complex interacting network

In this paper, we model the excitation energy transfer (EET) of photosystem I (PSI) of the common pea plant Pisum sativum as a complex interacting network. The magnitude of the link energy transfer between nodes/chromophores is computed by Forster resonant energy transfer (FRET) using the pairwise physical distances between chromophores from the PDB 5L8R (Protein Data Bank). We measure the global PSI network EET efficiency adopting well-known network theory indicators: the network efficiency (Eff) and the largest connected component (LCC). We also account the number of connected nodes/chromophores to P700 (CN), a new ad hoc measure we introduce here to indicate how many nodes in the network can actually transfer energy to the P700 reaction centre. We find that when progressively removing the weak links of lower EET, the Eff decreases, while the EET paths integrity (LCC and CN) is still preserved. This finding would show that the PSI is a resilient system owning a large window of functioning feasibility and it is completely impaired only when removing most of the network links. From the study of different types of chromophore, we propose different primary functions within the PSI system: chlorophyll a (CLA) molecules are the central nodes in the EET process, while other chromophore types have different primary functions. Furthermore, we perform nodes removal simulations to understand how the nodes/chromophores malfunctioning may affect PSI functioning. We discover that the removal of the CLA triggers the fastest decrease in the Eff, confirming that CAL is the main contributors to the high EET efficiency. Our outcomes open new perspectives of research, such comparing the PSI energy transfer efficiency of different natural and agricultural plant species and investigating the light-harvesting mechanisms of artificial photosynthesis both in plant agriculture and in the field of solar energy applications.