Bilateral Filtering Graph Convolutional Network for Multi-relational Social Recommendation in the Power-law Networks

In recent years, advances in Graph Convolutional Networks (GCNs) have given new insights into the development of social recommendation. However, many existing GCN-based social recommendation methods often directly apply GCN to capture user-item and user-user interactions, which probably have two main limitations: (a) Due to the power-law property of the degree distribution, the vanilla GCN with static normalized adjacency matrix has limitations in learning node representations, especially for the long-tail nodes; (b) multi-typed social relationships between users that are ubiquitous in the real world are rarely considered. In this article, we propose a novel Bilateral Filtering Heterogeneous Attention Network (BFHAN), which improves long-tail node representations and leverages multi-typed social relationships between user nodes. First, we propose a novel graph convolutional filter for the user-item bipartite network and extend it to the user-user homogeneous network. Further, we theoretically analyze the correlation between the convergence values of different graph convolutional filters and node degrees after stacking multiple layers. Second, we model multi-relational social interactions between users as the multiplex network and further propose a multiplex attention network to capture distinctive inter-layer influences for user representations. Last but not least, the experimental results demonstrate that our proposed method outperforms several state-of-the-art GCN-based methods for social recommendation tasks.

Research themes of family and community physicians in Brazil

Family and community medicine is a specialty dedicated to primary care, the cornerstone of effective health systems. Research capacity in primary care varies worldwide, and bibliographic databases such as MEDLINE, Scopus and Web of Science do not index most primary care research coming from Latin America. Our objective was to investigate the research themes of family and community physicians in Brazil, and to correlate the articles' research themes with their authors' trajectories in postgraduate education. For that, we compiled a national list of family and community physicians, retrieved their curricula from the Lattes Platform, compiled a list of journal articles, and obtained their keywords from LILACS and MEDLINE. Treating journal articles and their keywords as the two node types in a bipartite network, we derived research themes using the dual-projection algorithm, combining the Leiden algorithm with hierarchical clustering. We found two research themes to be the largest, most developed, and most central ones: human health and primary care. Authors with a master's or PhD in collective health (public health, epidemiology, and social sciences and humanities in health) were as likely as those with no postgraduate degree to publish articles on primary care. On the other hand, authors with a postgraduate degree in medicine were more likely to publish articles on human health. After discussing the findings in light of previous research and methodological aspects, we conclude there's a relative divide between primary care and clinical research, and the highlight policy implications.

Capturing a Comprehensive Picture of Biological Events From Adverse Outcome Pathways in the Drug Exposome

Background: The chemical part of the exposome, including drugs, may explain the increase of health effects with outcomes such as infertility, allergies, metabolic disorders, which cannot be only explained by the genetic changes. To better understand how drug exposure can impact human health, the concepts of adverse outcome pathways (AOPs) and AOP networks (AONs), which are representations of causally linked events at different biological levels leading to adverse health, could be used for drug safety assessment.Methods: To explore the action of drugs across multiple scales of the biological organization, we investigated the use of a network-based approach in the known AOP space. Considering the drugs and their associations to biological events, such as molecular initiating event and key event, a bipartite network was developed. This bipartite network was projected into a monopartite network capturing the event–event linkages. Nevertheless, such transformation of a bipartite network to a monopartite network had a huge risk of information loss. A way to solve this problem is to quantify the network reduction. We calculated two scoring systems, one measuring the uncertainty and a second one describing the loss of coverage on the developed event–event network to better investigate events from AOPs linked to drugs.Results: This AON analysis allowed us to identify biological events that are highly connected to drugs, such as events involving nuclear receptors (ER, AR, and PXR/SXR). Furthermore, we observed that the number of events involved in a linkage pattern with drugs is a key factor that influences information loss during monopartite network projection. Such scores have the potential to quantify the uncertainty of an event involved in an AON, and could be valuable for the weight of evidence assessment of AOPs. A case study related to infertility, more specifically to “decrease, male agenital distance” is presented.Conclusion: This study highlights that computational approaches based on network science may help to understand the complexity of drug health effects, with the aim to support drug safety assessment.

Comparing alternatives to the fixed degree sequence model for extracting the backbone of bipartite projections

Projections of bipartite or two-mode networks capture co-occurrences, and are used in diverse fields (e.g., ecology, economics, bibliometrics, politics) to represent unipartite networks. A key challenge in analyzing such networks is determining whether an observed number of co-occurrences between two nodes is significant, and therefore whether an edge exists between them. One approach, the fixed degree sequence model (FDSM), evaluates the significance of an edge’s weight by comparison to a null model in which the degree sequences of the original bipartite network are fixed. Although the FDSM is an intuitive null model, it is computationally expensive because it requires Monte Carlo simulation to estimate each edge’s p value, and therefore is impractical for large projections. In this paper, we explore four potential alternatives to FDSM: fixed fill model, fixed row model, fixed column model, and stochastic degree sequence model (SDSM). We compare these models to FDSM in terms of accuracy, speed, statistical power, similarity, and ability to recover known communities. We find that the computationally-fast SDSM offers a statistically conservative but close approximation of the computationally-impractical FDSM under a wide range of conditions, and that it correctly recovers a known community structure even when the signal is weak. Therefore, although each backbone model may have particular applications, we recommend SDSM for extracting the backbone of bipartite projections when FDSM is impractical.

Global Studies of the Host-Parasite Relationships between Ectoparasitic Mites of the Family Syringophilidae and Birds of the Order Columbiformes

The quill mites belonging to the family Syringophilidae (Acari: Prostigmata: Cheyletoidea) are obligate ectoparasites of birds. They inhabit different types of the quills, where they spend their whole life cycle. In this paper, we conducted a global study of syringophilid mites associated with columbiform birds. We examined 772 pigeon and dove individuals belonging to 112 species (35% world fauna) from all zoogeographical regions (except Madagascan) where Columbiformes occur. We measured the prevalence (IP) and the confidence interval (CI) for all infested host species. IP ranges between 4.2 and 66.7 (CI 0.2–100). We applied a bipartite analysis to determine host–parasite interaction, network indices, and host specificity on species and whole network levels. The Syringophilidae–Columbiformes network was composed of 25 mite species and 65 host species. The bipartite network was characterized by a high network level specialization H2′ = 0.93, high nestedness N = 0.908, connectance C = 0.90, and high modularity Q = 0.83, with 20 modules. Moreover, we reconstructed the phylogeny of the quill mites associated with columbiform birds on the generic level. Analysis shows two distinct clades: Meitingsunes + Psittaciphilus, and Peristerophila + Terratosyringophilus.

Local interactions and homophily effects in actor collaboration networks for urban resilience governance

Understanding actor collaboration networks and their evolution is essential to promoting collective action in resilience planning and management of interdependent infrastructure systems. Local interactions and choice homophily are two important network evolution mechanisms. Network motifs encode the information of network formation, configuration, and the local structure. Homophily effects, on the other hand, capture whether the network configurations have significant correlations with node properties. The objective of this paper is to explore the extent to which local interactions and homophily effects influence actor collaboration in resilience planning and management of interdependent infrastructure systems. We mapped bipartite actor collaboration network based on a post-Hurricane Harvey stakeholder survey that revealed actor collaborations for hazard mitigation. We examined seven bipartite network motifs for the mapped collaboration network and compared the mapped network to simulated random models with same degree distributions. Then we examined whether the network configurations had significant statistics for node properties using exponential random graph models. The results provide insights about the two mechanisms—local interactions and homophily effect—influencing the formation of actor collaboration in resilience planning and management of interdependent urban systems. The findings have implications for improving network cohesion and actor collaborations from diverse urban sectors.

Relationship Between Drug Targets and Drug-Signature Networks: A Network-Based Genome-Wide Landscape

Drugs produce pharmaceutical and adverse effects that arise from the complex relationship between drug targets and signatures; by considering such relationships, we can begin to understand the cellular mechanisms of drugs. In this study, we selected 463 genes from the DSigDB database corresponding to targets and signatures for 382 FDA-approved drugs with both protein binding information for a drug-target score (KDTN, i.e., the degree to which the protein encoded by the gene binds to a number of drugs) and microarray signature information for a drug-sensitive score (KDSN, i.e., the degree to which gene expression is stimulated by the drug). Accordingly, we constructed two drug–gene bipartite network models, a drug-target network and drug-signature network, which were merged into a multidimensional model. Analysis revealed that the KDTN and KDSN were in mutually exclusive and reciprocal relationships in terms of their biological network structure and gene function. A symmetric balance between the KDTN and KDSN of genes facilitates the possibility of therapeutic drug effects in living organisms. These results provide new insights into the relationship between drugs and genes, specifically drug targets and drug signatures.

An Effective Approach for Modular Community Detection in Bipartite Network Based on Integrating Rider with Harris Hawks Optimization Algorithms

The strenuous mining and arduous discovery of the concealed community structure in complex networks has received tremendous attention by the research community and is a trending domain in the multifaceted network as it not only reveals details about the hierarchical structure of multifaceted network but also assists in better understanding of the core functions of the network and subsequently information recommendation. The bipartite networks belong to the multifaceted network whose nodes can be divided into a dissimilar node-set so that no edges assist between the vertices. Even though the discovery of communities in one-mode network is briefly studied, community detection in bipartite networks is not studied. In this paper, we propose a novel Rider-Harris Hawks Optimization (RHHO) algorithm for community detection in a bipartite network through node similarity. The proposed RHHO is developed by the integration of the Rider Optimization (RO) algorithm with the Harris Hawks Optimization (HHO) algorithm. Moreover, a new evaluation metric, i.e., h-Tversky Index (h-TI), is also proposed for computing node similarity and fitness is newly devised considering modularity. The goal of modularity is to quantify the goodness of a specific division of network to evaluate the accuracy of the proposed community detection. The quantitative assessment of the proposed approach, as well as thorough comparative evaluation, was meticulously conducted in terms of fitness and modularity over the citation networks datasets (cit-HepPh and cit-HepTh) and bipartite network datasets (Movie Lens 100 K and American Revolution datasets). The performance was analyzed for 250 iterations of the simulation experiments. Experimental results have shown that the proposed method demonstrated a maximal fitness of 0.74353 and maximal modularity of 0.77433, outperforming the state-of-the-art approaches, including h-index-based link prediction, such as Multiagent Genetic Algorithm (MAGA), Genetic Algorithm (GA), Memetic Algorithm for Community Detection in Bipartite Networks (MATMCD-BN), and HHO.

Linking host plants to damage types in the fossil record of insect herbivory

Studies of insect herbivory on fossilized leaves tend to focus on a few, relatively simple metrics that are agnostic to the distribution of insect damage types among host plants. More complex metrics that link particular damage types to particular host plants have the potential to address additional ecological questions, but such metrics can be biased by sampling incompleteness due to the difficulty of distinguishing the true absence of a particular interaction from the failure to detect it---a challenge that has been raised in the ecological literature. We evaluate a range of methods for characterizing the relationships between damage types and host plants by performing resampling and subsampling exercises on a variety of datasets. We found that the components of beta diversity provide a more valid, reliable, and interpretable method for comparing component communities than do bipartite network metrics. We found the rarefaction of interactions to be a valid, reliable, and interpretable method for comparing compound communities. Both of these methods avoid the potential pitfalls of multiple comparisons. Lastly, we found that the host specificity of individual damage types is challenging to assess. Whereas some previously used methods are sufficiently biased by sampling incompleteness to be inappropriate for fossil herbivory data, alternatives exist that are perfectly suitable for fossil datasets with sufficient sample coverage.