Heterogeneous Interaction Networks Research Articles

Discovering Microbe-disease Associations with Weighted Graph Convolution Networks and Taxonomy Common Tree

Background: Microbe-disease associations are integral to understanding complex diseases and their screening procedures. Objective: While numerous computational methods have been developed to detect these associations, their performance remains limited due to inadequate utilization of weighted inherent similarities and microbial taxonomy hierarchy. To address this limitation, we have introduced WTHMDA (weighted taxonomic heterogeneous network-based microbe-disease association), a novel deep learning framework. Methods: WTHMDA combines a weighted graph convolution network and the microbial taxonomy common tree to predict microbe-disease associations effectively. The framework extracts multiple microbe similarities from the taxonomy common tree, facilitating the construction of a microbe- disease heterogeneous interaction network. Utilizing a weighted DeepWalk algorithm, node embeddings in the network incorporate weight information from the similarities. Subsequently, a deep neural network (DNN) model accurately predicts microbe-disease associations based on this interaction network. Results: Extensive experiments on multiple datasets and case studies demonstrate WTHMDA's superiority over existing approaches, particularly in predicting unknown associations. Conclusion: Our proposed method offers a new strategy for discovering microbe-disease linkages, showcasing remarkable performance and enhancing the feasibility of identifying disease risk.

A Novel Method for Mining Regulatory sRNAs Related to Rice Resistance Against Blast Fungus from Multi-Omics Data

Background: Due to infection by the rice blast fungus, rice, a major global staple, faces yield challenges. While chemical control methods are common, their environmental and economic costs are growing concerns. Traditional biological experiments are also inefficient for exploring resistance genes. Therefore, understanding the interaction between rice and the rice blast fungus is urgent and important. Objective: This study aims to use multi-omics data to uncover key elements in rice's defense against rice blast fungus Magnaporthe oryzae. We built a detailed, multi-layered heterogeneous interaction network, employing an innovative graph embedding feature with a cross-layer random walk algorithm to identify crucial crucial resistance factors.This could inform strategies for enhancing disease resistance in rice. objective: This study aims to use multi-omics data to uncover key elements in rice's defense against rice blast fungus Magnaporthe oryzae. We built a detailed, multi-layered heterogeneous interaction network, employing an innovative graph embedding feature with a cross-layer random walk algorithm, to identify crucial crucial resistance factors. This could inform strategies for enhancing disease resistance in rice. Methods: We integrated genomics, transcriptomics, and proteomics data on Magnaporthe oryzae infecting rice. This multi-omics data was used to construct a multi-layer heterogeneous network.An advanced graph embedding algorithm (BINE) provided rich vector representations of network nodes. A multi-layer network walking algorithm was then used to analyze the network and identify key regulatory small RNA (sRNAs) in rice. Results: Node similarity rankings allowed us to identify significant regulatory sRNAs in rice that are integral to disease resistance. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses further revealed their roles in biological processes and key metabolic pathways.Our integrative method precisely and efficiently identified these crucial elements, offering a valuable systems biology tool. Conclusion: By integrating multi-omics data with computational analysis, this study reveals key regulatory sRNAs in rice's disease resistance mechanism. These findings enhance our understanding of rice disease resistance and provide genetic resources for breeding disease-resistant rice. Despite limitations in sRNA functional interpretation, this research demonstrates the power of applying multi- omics data to address complex biological problems.

Identifying Autophagy-Associated Proteins and Chemicals with a Random Walk-Based Method within Heterogeneous Interaction Network.

Autophagy is instrumental in various health conditions, including cancer, aging, and infections. Therefore, examining proteins and compounds associated with autophagy is paramount to understanding cellular biology and the origins of diseases, paving the way for potential therapeutic and disease prediction strategies. However, the complexity of autophagy, its intersection with other cellular pathways, and the challenges in monitoring autophagic activity make the experimental identification of these elements arduous. In this study, autophagy-related proteins and chemicals were catalogued on the basis of Human Autophagy-dedicated Database. These entities were mapped to their respective PubChem identifications (IDs) for chemicals and Ensembl IDs for proteins, yielding 563 chemicals and 779 proteins. A network comprising protein-protein, protein-chemical, and chemical-chemical interactions was probed employing the Random-Walk-with-Restart algorithm using the aforementioned proteins and chemicals as seed nodes to unearth additional autophagy-associated proteins and chemicals. Screening tests were performed to exclude proteins and chemicals with minimal autophagy associations. A total of 88 inferred proteins and 50 inferred chemicals of high autophagy relevance were identified. Certain entities, such as the chemical prostaglandin E2 (PGE2), which is recognized for modulating cell death-induced inflammatory responses during pathogen invasion, and the protein G Protein Subunit Alpha I1 (GNAI1), implicated in ether lipid metabolism influencing a range of cellular processes including autophagy, were associated with autophagy. The discovery of novel autophagy-associated proteins and chemicals is of vital importance because it enhances the understanding of autophagy, provides potential therapeutic targets, and fosters the development of innovative therapeutic strategies and interventions.

Exploring multi-relational spatial interaction imputation with distance-decay effects

ABSTRACT Spatial interaction imputation aims to compensate for missing stable connections in geographical space, bolstering interaction network integrity and accuracy. Graph neural networks excel in graph-structured interaction data. However, existing research often focuses on homogeneous networks, neglecting the impact of heterogeneous interaction relationships influenced by distance decay on interaction imputation. Neglecting edge heterogeneity constrains the ability to effectively model the network structure, consequently leading to suboptimal performance in interaction imputation. This study introduces an interaction imputation graph convolutional network model. It constructs a heterogeneous interaction network with multi-distance relationships, considering distance decay. The model performs graph embedding based on interaction relationships between nodes. It comprehensively incorporates multiple interaction modes, topological structures, and node attributes to enhance spatial interaction imputation accuracy. Empirically validated using Beijing taxi travel data, our model outperforms existing models, improving imputation accuracy by approximately 8.70%. Our model consistently maintains superior accuracy in interaction networks of various sizes, demonstrating the stable superiority of our model. We also demonstrated that the variation in the number of interaction relationships affects imputation accuracy. A reasonable number of relationships and a larger feature dimension of geographical units yield better interaction imputation results.

Heterogeneous context interaction network for vehicle re-identification

Capturing global and subtle discriminative information using attention mechanisms is essential to address the challenge of inter-class high similarity for vehicle re-identification (Re-ID) task. Mixing self-information of nodes or modeling context based on pairwise dependencies between nodes are the core ideas of current advanced attention mechanisms. This paper aims to explore how to utilize both dependency context and self-context in an efficient way to facilitate attention to learn more effectively. We propose a heterogeneous context interaction (HCI) attention mechanism that infers the weights of nodes from the interactions of global dependency contexts and local self-contexts to enhance the effect of attention learning. To reduce computational complexity, global dependency contexts are modeled by aggregating number-compressed pairwise dependencies, and the interactions of heterogeneous contexts are restricted to a certain range. Based on this mechanism, we propose a heterogeneous context interaction network (HCI-Net), which uses channel heterogeneous context interaction module (CHCI) and spatial heterogeneous context interaction module (SHCI), and introduces a rigid partitioning strategy to extract important global and fine-grained features. In addition, we design a non-similarity constraint (NSC) that forces the HCI-Net to learn diverse subtle discriminative information. The experiment results on two large datasets, VeRi-776 and VehicleID, show that our proposed HCI-Net achieves the state-of-the-art performance. In particular, the mean average precision (mAP) reaches 83.8% on VeRi-776 dataset.

A Relationship Prediction Method for Magnaporthe oryzae-Rice Multi-Omics Data Based on WGCNA and Graph Autoencoder.

Magnaporthe oryzae Oryzae (MoO) pathotype is a devastating fungal pathogen of rice; however, its pathogenic mechanism remains poorly understood. The current research is primarily focused on single-omics data, which is insufficient to capture the complex cross-kingdom regulatory interactions between MoO and rice. To address this limitation, we proposed a novel method called Weighted Gene Autoencoder Multi-Omics Relationship Prediction (WGAEMRP), which combines weighted gene co-expression network analysis (WGCNA) and graph autoencoder to predict the relationship between MoO-rice multi-omics data. We applied WGAEMRP to construct a MoO-rice multi-omics heterogeneous interaction network, which identified 18 MoO small RNAs (sRNAs), 17 rice genes, 26 rice mRNAs, and 28 rice proteins among the key biomolecules. Most of the mined functional modules and enriched pathways were related to gene expression, protein composition, transportation, and metabolic processes, reflecting the infection mechanism of MoO. Compared to previous studies, WGAEMRP significantly improves the efficiency and accuracy of multi-omics data integration and analysis. This approach lays out a solid data foundation for studying the biological process of MoO infecting rice, refining the regulatory network of pathogenic markers, and providing new insights for developing disease-resistant rice varieties.

PTBGRP: predicting phage-bacteria interactions with graph representation learning on microbial heterogeneous information network.

Identifying the potential bacteriophages (phage) candidate to treat bacterial infections plays an essential role in the research of human pathogens. Computational approaches are recognized as a valid way to predict bacteria and target phages. However, most of the current methods only utilize lower-order biological information without considering the higher-order connectivity patterns, which helps to improve the predictive accuracy. Therefore, we developed a novel microbial heterogeneous interaction network (MHIN)-based model called PTBGRP to predict new phages for bacterial hosts. Specifically, PTBGRP first constructs an MHIN by integrating phage-bacteria interaction (PBI) and six bacteria-bacteria interaction networks with their biological attributes. Then, different representation learning methods are deployed to extract higher-level biological features and lower-level topological features from MHIN. Finally, PTBGRP employs a deep neural network as the classifier to predict unknown PBI pairs based on the fused biological information. Experiment results demonstrated that PTBGRP achieves the best performance on the corresponding ESKAPE pathogens and PBI dataset when compared with state-of-art methods. In addition, case studies of Klebsiella pneumoniae and Staphylococcus aureus further indicate that the consideration of rich heterogeneous information enables PTBGRP to accurately predict PBI from a more comprehensive perspective. The webserver of the PTBGRP predictor is freely available at http://120.77.11.78/PTBGRP/.

ASpediaFI: Functional Interaction Analysis of Alternative Splicing Events

Alternative splicing (AS) regulates biological processes governing phenotypes and diseases. Differential AS (DAS) gene test methods have been developed to investigate important exonic expression from high-throughput datasets. However, the DAS events extracted using statistical tests are insufficient to delineate relevant biological processes. In this study, we developed a novel application, Alternative Splicing Encyclopedia: Functional Interaction (ASpediaFI), to systemically identify DAS events and co-regulated genes and pathways. ASpediaFI establishes a heterogeneous interaction network of genes and their feature nodes (i.e., AS events and pathways) connected by co-expression or pathway gene set knowledge. Next, ASpediaFI explores the interaction network using the random walk with restart algorithm and interrogates the proximity from a query gene set. Finally, ASpediaFI extracts significant AS events, genes, and pathways. To evaluate the performance of our method, we simulated RNA sequencing (RNA-seq) datasets to consider various conditions of sequencing depth and sample size. The performance was compared with that of other methods. Additionally, we analyzed three public datasets of cancer patients or cell lines to evaluate how well ASpediaFI detects biologically relevant candidates. ASpediaFI exhibits strong performance in both simulated and public datasets. Our integrative approach reveals that DAS events that recognize a global co-expression network and relevant pathways determine the functional importance of spliced genes in the subnetwork. ASpediaFI is publicly available at https://bioconductor.org/packages/ASpediaFI.

DTI-HeNE: a novel method for drug-target interaction prediction based on heterogeneous network embedding

BackgroundPrediction of the drug-target interaction (DTI) is a critical step in the drug repurposing process, which can effectively reduce the following workload for experimental verification of potential drugs’ properties. In recent studies, many machine-learning-based methods have been proposed to discover unknown interactions between drugs and protein targets. A recent trend is to use graph-based machine learning, e.g., graph embedding to extract features from drug-target networks and then predict new drug-target interactions. However, most of the graph embedding methods are not specifically designed for DTI predictions; thus, it is difficult for these methods to fully utilize the heterogeneous information of drugs and targets (e.g., the respective vertex features of drugs and targets and path-based interactive features between drugs and targets).ResultsWe propose a DTI prediction method DTI-HeNE (DTI based on Heterogeneous Network Embedding), which is specifically designed to cope with the bipartite DTI relations for generating high-quality embeddings of drug-target pairs. This method splits a heterogeneous DTI network into a bipartite DTI network, multiple drug homogeneous networks and target homogeneous networks, and extracts features from these sub-networks separately to better utilize the characteristics of bipartite DTI relations as well as the auxiliary similarity information related to drugs and targets. The features extracted from each sub-network are integrated using pathway information between these sub-networks to acquire new features, i.e., embedding vectors of drug-target pairs. Finally, these features are fed into a random forest (RF) model to predict novel DTIs.ConclusionsOur experimental results show that, the proposed DTI network embedding method can learn higher-quality features of heterogeneous drug-target interaction networks for novel DTIs discovery.

IdenPC-CAP: Identify protein complexes from weighted RNA-protein heterogeneous interaction networks using co-assemble partner relation.

Protein complexes play important roles in most cellular processes. The available genome-wide protein-protein interaction (PPI) data make it possible for computational methods identifying protein complexes from PPI networks. However, PPI datasets usually contain a large ratio of false positive noise. Moreover, different types of biomolecules in a living cell cooperate to form a union interaction network. Because previous computational methods focus only on PPIs ignoring other types of biomolecule interactions, their predicted protein complexes often contain many false positive proteins. In this study, we develop a novel computational method idenPC-CAP to identify protein complexes from the RNA-protein heterogeneous interaction network consisting of RNA-RNA interactions, RNA-protein interactions and PPIs. By considering interactions among proteins and RNAs, the new method reduces the ratio of false positive proteins in predicted protein complexes. The experimental results demonstrate that idenPC-CAP outperforms the other state-of-the-art methods in this field.

Effect of collective influence on the evolution of cooperation in evolutionary prisoner’s dilemma games

Node centrality plays an important role in many dynamical processes taking place on complex networks. In this work, we associate the individuals’ collective influence (CI) with their strategy-updating time scales to investigate how the diverse collective influence of individuals affects the evolution of cooperation in the evolutionary prisoner’s dilemma game on scale-free networks. With the combination of time scale mechanism which bridges the feedback between strategy-updating time scale and the performance of individuals, we find that influential cooperators locating at medium- or small-degrees are able to spread their behaviors among neighbors in a more efficient way than hubs with large-degrees. Hence, collective influence with proper path length can efficiently identify influencers and may promote the emergence of cooperation on heterogeneous interaction networks.

Behavioral Link Analytics on Heterogeneous Human Interaction Networks

For designing and modeling Artificial Intelligence (AI) systems in the area of human-machine interaction, suitable approaches for user modeling are important in order to both capture user characteristics. Using multimodal data, this can be performed from various perspectives. Specifically, for modeling user interactions in human interaction networks, appropriate approaches for capturing those interactions, as well as to analyze them in order to extract meaningful patterns are important. Specifically, for modeling user behavior for the respective AI systems, we can make use of diverse heterogeneous data sources. This paper investigates face-to-face as well as socio-spatial interaction networks for modeling user interactions from three perspectives: We analyze preferences and perceptions of human social interactions in relation to the interactions observed using wearable sensors, i. e., face-to-face as well as socio-spatial interactions fo the respective actors. For that, we investigate the correspondence of according networks, in order to identify conformance, exceptions, and anomalies. The analysis is performed on a real-world dataset capturing networks of proximity interactions coupled with self-report questionnaires about preferences and perception of those interactions. The different networks, and according perspectives then provide different options for user modeling and integration into AI systems modeling such user behavior.

Drug-target Interaction Prediction by Metapath2vec Node Embedding in Heterogeneous Network of Interactions

Drug discovery is a complicated, time-consuming and expensive process. The cost for each new molecular entity (NME) is estimated at $1.8 billion. Furthermore, for a new drug to be FDA approved it often takes nearly a decade and approximately 20 new drugs being approved by the US Food and Drug Administration (FDA) each year. Accurately predicting drug-target interactions (DTIs) by computational methods is an important area of drug research, which brings about a broad prospect for fast and low-risk drug development. By accurate prediction of drugs and targets interactions scientists can scale-down huge experimental space and reduce the costs and help to faster drug development as well as predicting the side effects and potential function of new drugs. Many approaches have been taken by researchers to solve DTI problem and enhance the accuracy of methods. State-of-the-art approaches are based on various techniques, such as deep learning methods-like stacked auto-encoder-, matrix factorization, network inference, and ensemble methods. In this work, we have taken a new approach based on node embedding in a heterogeneous interaction network to obtain the representation of each node in the interaction network and then use a binary classifier such as logistic regression to solve this prominent problem in the pharmaceutical industry. Most introduced network-based methods use a homogeneous network of interactions as their input data whereas in the real word problem there exist other informative networks to help to enhance the prediction and by considering the homogeneous networks we lose some precious network information. Hence, in this work, we have tried to work on the heterogeneous network and have improved the accuracy of methods in comparison to baseline methods.

POSSIBILIDADES DE CONVIVÊNCIA NA HETEROGENEIDADE: UMA EXPERIÊNCIA PELAS RUAS DA CIDADE

This study discusses an action developed by the Outreach Project Lambedores de Cultura, which aims to promote interaction with and in diversity and to establish heterogeneous networks, exploring the possibilities of the territory of the cities where the project takes place. The project is based on the guiding principles of Community Centers and Cooperative, which see in art and culture possibilities of meaningful interactions between people. The article refers to a Photographic Exploration Workshop held in Rio Claro / SP. The experiment used photography as a means to explore and occupy different city spaces, thus seeking to foster and increase heterogeneous interaction networks. The weekly workshops were opened to the community and included the participation of segments of the population segregated from common city life, due to several social markers. The workshops promoted new spaces for socialization as well as the appropriation of public spaces.

Kinetic Modelling of Competition and Depletion of Shared miRNAs by Competing Endogenous RNAs.

Non-coding RNAs play a key role in the post-transcriptional regulation of mRNA translation and turnover in eukaryotes. miRNAs, in particular, interact with their target RNAs through protein-mediated, sequence-specific binding, giving rise to extended and highly heterogeneous miRNA-RNA interaction networks. Within such networks, competition to bind miRNAs can generate an effective positive coupling between their targets. Competing endogenous RNAs (ceRNAs) can in turn regulate each other through miRNA-mediated crosstalk. Albeit potentially weak, ceRNA interactions can occur both dynamically, affecting, e.g., the regulatory clock, and at stationarity, in which case ceRNA networks as a whole can be implicated in the composition of the cell's proteome. Many features of ceRNA interactions, including the conditions under which they become significant, can be unraveled by mathematical and in silico models. We review the understanding of the ceRNA effect obtained within such frameworks, focusing on the methods employed to quantify it, its role in the processing of gene expression noise, and how network topology can determine its reach.

Arena-Idb: a platform to build human non-coding RNA interaction networks

BackgroundHigh throughput technologies have provided the scientific community an unprecedented opportunity for large-scale analysis of genomes. Non-coding RNAs (ncRNAs), for a long time believed to be non-functional, are emerging as one of the most important and large family of gene regulators and key elements for genome maintenance. Functional studies have been able to assign to ncRNAs a wide spectrum of functions in primary biological processes, and for this reason they are assuming a growing importance as a potential new family of cancer therapeutic targets. Nevertheless, the number of functionally characterized ncRNAs is still too poor if compared to the number of new discovered ncRNAs. Thus platforms able to merge information from available resources addressing data integration issues are necessary and still insufficient to elucidate ncRNAs biological roles.ResultsIn this paper, we describe a platform called Arena-Idb for the retrieval of comprehensive and non-redundant annotated ncRNAs interactions. Arena-Idb provides a framework for network reconstruction of ncRNA heterogeneous interactions (i.e., with other type of molecules) and relationships with human diseases which guide the integration of data, extracted from different sources, via mapping of entities and minimization of ambiguity.ConclusionsArena-Idb provides a schema and a visualization system to integrate ncRNA interactions that assists in discovering ncRNA functions through the extraction of heterogeneous interaction networks. The Arena-Idb is available at http://arenaidb.ba.itb.cnr.it

Plant geographic phenotypic variation drives diversification in its associated community of a phytophagous insect and its parasitoids

BackgroundWhile the communities constituted by phytophageous insects and their parasites may represent half of all terrestrial animal species, understanding their diversification remains a major challenge. A neglected idea is that geographic phenotypic variation in a host plant may lead to heterogeneous evolutionary responses of the different members of the associated communities. This could result in diversification on a host plant by ecological speciation in some species, leading to geographic variation in community composition. In this study we investigated geographic variation of inflorescence receptacle size in a plant, Ficus hirta, and how the hymenopteran community feeding in the inflorescences has responded. Our predictions were:Inflorescence size variation affects wasp species differently depending on how they access oviposition sites.In some affected lineages of wasps, we may observe vicariant, parapatric species adapted to different inflorescence sizes.ResultsWe show that fig (the enclosed inflorescence of Ficus) wall thickness varies geographically. The fig-entering pollinating wasp was not affected, while the parasites ovipositing through the fig wall were. Two parapatric species of Philotrypesis, exhibiting strikingly different ovipositor lengths, were recorded. One species of Sycoscapter was also present, and it was restricted, like the shorter-ovipositor Philotrypesis, to the geographic zone where fig walls were thinner.ConclusionsPrevious work on fig wasps suggested that parapatric geographic ranges among congenerics were due to adaptation to variation in abiotic factors, complemented by interspecific competition. Our results show that parapatric ranges may also result from adaptation to variation in biotic factors. Within an insect community, differences among species in their response to geographic phenotypic variation of their host plant may result in geographically heterogeneous community structure. Such heterogeneity leads to heterogeneous interaction networks among sites. Our results support the hypothesis that plant geographic phenotypic variation can be a driver of diversification in associated insect communities, and can complement other diversification processes.

DeepNF: deep network fusion for protein function prediction.

MotivationThe prevalence of high-throughput experimental methods has resulted in an abundance of large-scale molecular and functional interaction networks. The connectivity of these networks provides a rich source of information for inferring functional annotations for genes and proteins. An important challenge has been to develop methods for combining these heterogeneous networks to extract useful protein feature representations for function prediction. Most of the existing approaches for network integration use shallow models that encounter difficulty in capturing complex and highly non-linear network structures. Thus, we propose deepNF, a network fusion method based on Multimodal Deep Autoencoders to extract high-level features of proteins from multiple heterogeneous interaction networks.ResultsWe apply this method to combine STRING networks to construct a common low-dimensional representation containing high-level protein features. We use separate layers for different network types in the early stages of the multimodal autoencoder, later connecting all the layers into a single bottleneck layer from which we extract features to predict protein function. We compare the cross-validation and temporal holdout predictive performance of our method with state-of-the-art methods, including the recently proposed method Mashup. Our results show that our method outperforms previous methods for both human and yeast STRING networks. We also show substantial improvement in the performance of our method in predicting gene ontology terms of varying type and specificity.Availability and implementation deepNF is freely available at: https://github.com/VGligorijevic/deepNF.Supplementary information Supplementary data are available at Bioinformatics online.

EVOLUTIONARY GAMES ON VISIBILITY GRAPHS

We show that time series of different complexities can be transformed into networks that host individuals playing evolutionary games. The irregularity of the time series is thereby faithfully reflected in the fraction of cooperators surviving the evolutionary process, thus effectively linking time series with evolutionary games. Pivotal to the linkage is a simple visibility algorithm that transforms time series into networks. More specifically, periodic series yield regular networks, chaotic series yield random networks, while fractal series yield scale-free networks. As an example, we use a chaotic time series from the Logistic map and a fractal time series of Brownian motion, yielding an interaction network with an exponential and a power-law degree distribution, respectively. By employing the prisoner's dilemma and the snowdrift game, we demonstrate that such heterogeneous interaction networks facilitate the evolution of cooperation if compared to the traditional square lattice topology. Due to the simplicity of the employed methodology, newcomers with a basic command of nonlinear dynamics or stochastic processes can become easily acquainted with evolutionary games, and moreover, integrate these interesting and vibrant subfields of physics more effectively into their research.

Resolving the structure of interactomes with hierarchical agglomerative clustering

BackgroundGraphs provide a natural framework for visualizing and analyzing networks of many types, including biological networks. Network clustering is a valuable approach for summarizing the structure in large networks, for predicting unobserved interactions, and for predicting functional annotations. Many current clustering algorithms suffer from a common set of limitations: poor resolution of top-level clusters; over-splitting of bottom-level clusters; requirements to pre-define the number of clusters prior to analysis; and an inability to jointly cluster over multiple interaction types.ResultsA new algorithm, Hierarchical Agglomerative Clustering (HAC), is developed for fast clustering of heterogeneous interaction networks. This algorithm uses maximum likelihood to drive the inference of a hierarchical stochastic block model for network structure. Bayesian model selection provides a principled method for collapsing the fine-structure within the smallest groups, and for identifying the top-level groups within a network. Model scores are additive over independent interaction types, providing a direct route for simultaneous analysis of multiple interaction types. In addition to inferring network structure, this algorithm generates link predictions that with cross-validation provide a quantitative assessment of performance for real-world examples.ConclusionsWhen applied to genome-scale data sets representing several organisms and interaction types, HAC provides the overall best performance in link prediction when compared with other clustering methods and with model-free graph diffusion kernels. Investigation of performance on genome-scale yeast protein interactions reveals roughly 100 top-level clusters, with a long-tailed distribution of cluster sizes. These are in turn partitioned into 1000 fine-level clusters containing 5 proteins on average, again with a long-tailed size distribution. Top-level clusters correspond to broad biological processes, whereas fine-level clusters correspond to discrete complexes. Surprisingly, link prediction based on joint clustering of physical and genetic interactions performs worse than predictions based on individual data sets, suggesting a lack of synergy in current high-throughput data.