Ranking candidate disease genes from gene expression and protein interaction: a Katz-centrality based approach

Ensemble decision of local similarity indices on the biological network for disease related gene prediction

Link prediction (LP) is a task for the identification of potential, missing and spurious links in complex networks. Protein-protein interaction (PPI) networks are important for understanding the underlying biological mechanisms of diseases. Many complex networks have been constructed using LP methods; however, there are a limited number of studies that focus on disease-related gene predictions and evaluate these genes using various evaluation criteria. The main objective of the study is to investigate the effect of a simple ensemble method in disease related gene predictions. Local similarity indices (LSIs) based disease related gene predictions were integrated by a simple ensemble decision method, simple majority voting (SMV), on the PPI network to detect accurate disease related genes. Human PPI network was utilized to discover potential disease related genes using four LSIs for the gene prediction. LSIs discovered potential links between disease related genes, which were obtained from OMIM database for gastric, colorectal, breast, prostate and lung cancers. LSIs based disease related genes were ranked due to their LSI scores in descending order for retrieving the top 10, 50 and 100 disease related genes. SMV integrated four LSIs based predictions to obtain SMV based the top 10, 50 and 100 disease related genes. The performance of LSIs based and SMV based genes were evaluated separately by employing overlap analyses, which were performed with GeneCard disease-gene relation dataset and Gene Ontology (GO) terms. The GO-terms were used for biological assessment for the inferred gene lists by LSIs and SMV on all cancer types. Adamic-Adar (AA), Resource Allocation Index (RAI), and SMV based gene lists are generally achieved good performance results on all cancers in both overlap analyses. SMV also outperformed on breast cancer data. The increment in the selection of the number of the top ranked disease related genes also enhanced the performance results of SMV.

Blueprint of differentially expressed genes reveals the dynamic gene expression landscape and the gender biases in long COVID

BACKGROUND

Long COVID has appeared as a significant global health issue and is an extra burden to the healthcare system. It affects a considerable number of people throughout the globe. However, substantial research gaps have been noted in understanding the mechanism and genomic landscape during the long COVID infection. A study has aimed to identify the differentially expressed genes (DEGs) in long COVID patients to fill the gap.

METHODS

We used the RNA-seq GEO dataset acquired through the GPL20301 Illumina HiSeq 4000 platform. The dataset contains 36 human samples derived from PBMC (Peripheral blood mononuclear cells). Thirty-six human samples contain 13 non-long COVID individuals' samples and 23 long COVID individuals' samples, considered the first direction analysis. Here, we performed two-direction analyses. In the second direction analysis, we divided the dataset gender-wise into four groups: the non-long COVID male group, the long COVID male group, the non-long COVID female group, and the long COVID female group.

RESULTS

In the first analysis, we found no gene expression. In the second analysis, we identified 250 DEGs. During the DEG profile analysis of the non-long COVID male group and the long COVID male group, we found three upregulated genes: IGHG2, IGHG4, and MIR8071-2. Similarly, the analysis of the non-long COVID female group and the long COVID female group reveals eight top-ranking genes. It also indicates the gender biases of differentially expressed genes among long COVID individuals. We found several DEGs involved in PPI and co-expression network formation. Similarly, cluster enrichment and gene list enrichment analysis were performed, suggesting several genes are involved in different biological pathways or processes.

CONCLUSIONS

This study will help better understand the gene expression landscape in long COVID. However, it might help the discovery and development of therapeutics for long COVID.

A QUBO formulation for top-τ eigencentrality nodes

The efficient calculation of the centrality or "hierarchy" of nodes in a network has gained great relevance in recent years due to the generation of large amounts of data. The eigenvector centrality (aka eigencentrality) is quickly becoming a good metric for centrality due to both its simplicity and fidelity. In this work we lay the foundations for solving the eigencentrality problem of ranking the importance of the nodes of a network with scores from the eigenvector of the network, using quantum computational paradigms such as quantum annealing and gate-based quantum computing. The problem is reformulated as a quadratic unconstrained binary optimization (QUBO) that can be solved on both quantum architectures. The results focus on correctly identifying a given number of the most important nodes in numerous networks given by the sparse vector solution of our QUBO formulation of the problem of identifying the top-τ highest eigencentrality nodes in a network on both the D-Wave and IBM quantum computers.

Scalable Non-Linear Graph Fusion for Prioritizing Cancer-Causing Genes

In the past few decades, both gene expression data and protein-protein interaction (PPI)networks have been extensively studied, due to their ability to depict important characteristics of disease-associated genes. In this regard, the paper presents a new gene prioritization algorithm to identify and prioritize cancer-causing genes, integrating judiciously the complementary information obtained from two data sources. The proposed algorithm selects disease-causing genes by maximizing the importance of selected genes and functional similarity among them. A new quantitative index is introduced to evaluate the importance of a gene. It considers whether a gene exhibits a differential expression pattern across sick and healthy individuals, and has a strong connectivity in the PPI network, which are the important characteristics of a potential biomarker. As disease-associated genes are expected to have similar expression profiles and topological structures, a scalable non-linear graph fusion technique, termed as ScaNGraF, is proposed to learn a disease-dependent functional similarity network from the co-expression and common neighbor based similarity networks. The proposed ScaNGraF, which is based on message passing algorithm, efficiently combines the shared and complementary information provided by different data sources with significantly lower computational cost. A new measure, termed as DiCoIN, is introduced to evaluate the quality of a learned affinity network. The performance of the proposed graph fusion technique and gene selection algorithm is extensively compared with that of some existing methods, using several cancer data sets.

Identifying Influential Nodes in Complex Networks Based on Multiple Local Attributes and Information Entropy

Identifying influential nodes in complex networks has attracted the attention of many researchers in recent years. However, due to the high time complexity, methods based on global attributes have become unsuitable for large-scale complex networks. In addition, compared with methods considering only a single attribute, considering multiple attributes can enhance the performance of the method used. Therefore, this paper proposes a new multiple local attributes-weighted centrality (LWC) based on information entropy, combining degree and clustering coefficient; both one-step and two-step neighborhood information are considered for evaluating the influence of nodes and identifying influential nodes in complex networks. Firstly, the influence of a node in a complex network is divided into direct influence and indirect influence. The degree and clustering coefficient are selected as direct influence measures. Secondly, based on the two direct influence measures, we define two indirect influence measures: two-hop degree and two-hop clustering coefficient. Then, the information entropy is used to weight the above four influence measures, and the LWC of each node is obtained by calculating the weighted sum of these measures. Finally, all the nodes are ranked based on the value of the LWC, and the influential nodes can be identified. The proposed LWC method is applied to identify influential nodes in four real-world networks and is compared with five well-known methods. The experimental results demonstrate the good performance of the proposed method on discrimination capability and accuracy.

DDMF: A Method for Mining Relatively Important Nodes Based on Distance Distribution and Multi-Index Fusion

In research on complex networks, mining relatively important nodes is a challenging and practical work. However, little research has been done on mining relatively important nodes in complex networks, and the existing relatively important node mining algorithms cannot take into account the indicators of both precision and applicability. Aiming at the scarcity of relatively important node mining algorithms and the limitations of existing algorithms, this paper proposes a relatively important node mining method based on distance distribution and multi-index fusion (DDMF). First, the distance distribution of each node is generated according to the shortest path between nodes in the network; then, the cosine similarity, Euclidean distance and relative entropy are fused, and the entropy weight method is used to calculate the weights of different indexes; Finally, by calculating the relative importance score of nodes in the network, the relatively important nodes are mined. Through verification and analysis on real network datasets in different fields, the results show that the DDMF method outperforms other relatively important node mining algorithms in precision, recall, and AUC value.

Determination of biomarkers from microarray data using graph neural network and spectral clustering

In bioinformatics, the rapid development of gene sequencing technology has produced an increasing amount of microarray data. This type of data shares the typical characteristics of small sample size and high feature dimensions. Searching for biomarkers from microarray data, which expression features of various diseases, is essential for the disease classification. feature selection has therefore became fundemental for the analysis of microarray data, which designs to remove irrelevant and redundant features. There are a large number of redundant features and irrelevant features in microarray data, which severely degrade the classification effectiveness. We propose an innovative feature selection method with the goal of obtaining feature dependencies from a priori knowledge and removing redundant features using spectral clustering. In this paper, the graph structure is firstly constructed by using the gene interaction network as a priori knowledge, and then a link prediction method based on graph neural network is proposed to enhance the graph structure data. Finally, a feature selection method based on spectral clustering is proposed to determine biomarkers. The classification accuracy on DLBCL and Prostate can be improved by 10.90% and 16.22% compared to traditional methods. Link prediction provides an average classification accuracy improvement of 1.96% and 1.31%, and is up to 16.98% higher than the published method. The results show that the proposed method can have full use of a priori knowledge to effectively select disease prediction biomarkers with high classification accuracy.

Stepwise target controllability identifies dysregulations of macrophage networks in multiple sclerosis

Identifying the nodes able to drive the state of a network is crucial to understand, and eventually control, biological systems. Despite recent advances, such identification remains difficult because of the huge number of equivalent controllable configurations, even in relatively simple networks. Based on the evidence that in many applications it is essential to test the ability of individual nodes to control a specific target subset, we develop a fast and principled method to identify controllable driver-target configurations in sparse and directed networks. We demonstrate our approach on simulated networks and experimental gene networks to characterize macrophage dysregulation in human subjects with multiple sclerosis.

Finding associations in a heterogeneous setting: statistical test for aberration enrichment

Most two-group statistical tests find broad patterns such as overall shifts in mean, median, or variance. These tests may not have enough power to detect effects in a small subset of samples, e.g., a drug that works well only on a few patients. We developed a novel statistical test targeting such effects relevant for clinical trials, biomarker discovery, feature selection, etc. We focused on finding meaningful associations in complex genetic diseases in gene expression, miRNA expression, and DNA methylation. Our test outperforms traditional statistical tests in simulated and experimental data and detects potentially disease-relevant genes with heterogeneous effects.

Microbes and complex diseases: from experimental results to computational models

Studies have shown that the number of microbes in humans is almost 10 times that of cells. These microbes have been proven to play an important role in a variety of physiological processes, such as enhancing immunity, improving the digestion of gastrointestinal tract and strengthening metabolic function. In addition, in recent years, more and more research results have indicated that there are close relationships between the emergence of the human noncommunicable diseases and microbes, which provides a novel insight for us to further understand the pathogenesis of the diseases. An in-depth study about the relationships between diseases and microbes will not only contribute to exploring new strategies for the diagnosis and treatment of diseases but also significantly heighten the efficiency of new drugs development. However, applying the methods of biological experimentation to reveal the microbe-disease associations is costly and inefficient. In recent years, more and more researchers have constructed multiple computational models to predict microbes that are potentially associated with diseases. Here, we start with a brief introduction of microbes and databases as well as web servers related to them. Then, we mainly introduce four kinds of computational models, including score function-based models, network algorithm-based models, machine learning-based models and experimental analysis-based models. Finally, we summarize the advantages as well as disadvantages of them and set the direction for the future work of revealing microbe-disease associations based on computational models. We firmly believe that computational models are expected to be important tools in large-scale predictions of disease-related microbes.

Unveiling new disease, pathway, and gene associations via multi-scale neural network

Diseases involve complex modifications to the cellular machinery. The gene expression profile of the affected cells contains characteristic patterns linked to a disease. Hence, new biological knowledge about a disease can be extracted from these profiles, improving our ability to diagnose and assess disease risks. This knowledge can be used for drug re-purposing, or by physicians to evaluate a patient’s condition and co-morbidity risk. Here, we consider differential gene expressions obtained by microarray technology for patients diagnosed with various diseases. Based on these data and cellular multi-scale organization, we aim at uncovering disease–disease, disease–gene and disease–pathway associations. We propose a neural network with structure based on the multi-scale organization of proteins in a cell into biological pathways. We show that this model is able to correctly predict the diagnosis for the majority of patients. Through the analysis of the trained model, we predict disease–disease, disease–pathway, and disease–gene associations and validate the predictions by comparisons to known interactions and literature search, proposing putative explanations for the predictions.

Network analysis of host-pathogen protein interactions in microbe induced cardiovascular diseases

Large-scale visualization and analysis of HPIs involved in microbial CVDs can provide crucial insights into the mechanisms of pathogenicity. The comparison of CVD associated HPIs with the entire set of HPIs can identify the pathways specific to CVDs. Therefore, topological properties of HPI networks in CVDs and all pathogens was studied using Cytoscape3.5.1. Ontology and pathway analysis were done using KOBAS 3.0. HPIs of Papilloma, Herpes, Influenza A virus as well as Yersinia pestis and Bacillus anthracis among bacteria were predominant in the whole (wHPI) and the CVD specific (cHPI) network. The central viral and secretory bacterial proteins were predicted virulent. The central viral proteins had higher number of interactions with host proteins in comparison with bacteria. Major fraction of central and essential host proteins interacts with central viral proteins. Alpha-synuclein, Ubiquitin ribosomal proteins, TATA-box-binding protein, and Polyubiquitin-C &B proteins were the top interacting proteins specific to CVDs. Signaling by NGF, Fc epsilon receptor, EGFR and ubiquitin mediated proteolysis were among the top enriched CVD specific pathways. DEXDc and HELICc were enriched host mimicry domains that may help in hijacking of cellular machinery by pathogens. This study provides a system level understanding of cardiac damage in microbe induced CVDs.

A network-based analysis for mining the risk pathways in glioblastoma

The most malignant type of brain tumour is glioblastoma multiforme (GBM). Patients with GBM often have a poor prognosis, as a result of incomplete or inaccurate diagnoses. Regulatory pathways have been demonstrated to serve important roles in complex human diseases. Therefore, deciphering these risk pathways may shed light on the molecular mechanisms underlying GBM progression. In the present study, differentially expressed genes and microRNAs (miRNAs) in a publicly available database were identified between normal and tumour samples. To determine the pathophysiology and molecular mechanisms underlying GBM, integrated network analysis was performed to mine GBM-specific risk pathways. Specifically, a GBM-specific regulatory network was constructed that integrated manually curated GBM-associated transcription and post-transcriptional data resources, including transcription factors and miRNAs. A total of 1,827 differentially expressed genes and 30 miRNAs were identified. The differentially expressed genes were significantly enriched in a number of immune response-associated functions. Based on the GBM-specific regulatory network, 15 risk regulatory pathways containing not only known regulators, but also potential novel targets that might be involved in tumourigenesis were identified. Network analysis provides a strategy for leveraging genomic data to identify potential oncogenic pathways and molecular targets for GBM.

Prediction of seed gene function in progressive diabetic neuropathy by a network-based inference method

Guilt by association (GBA) algorithm has been widely used to statistically predict gene functions, and network-based approach increases the confidence and veracity of identifying molecular signatures for diseases. This work proposed a network-based GBA method by integrating the GBA algorithm and network, to identify seed gene functions for progressive diabetic neuropathy (PDN). The inference of predicting seed gene functions comprised of three steps: i) Preparing gene lists and sets; ii) constructing a co-expression matrix (CEM) on gene lists by Spearman correlation coefficient (SCC) method and iii) predicting gene functions by GBA algorithm. Ultimately, seed gene functions were selected according to the area under the receiver operating characteristics curve (AUC) index. A total of 79 differentially expressed genes (DEGs) and 40 background gene ontology (GO) terms were regarded as gene lists and sets for the subsequent analyses, respectively. The predicted results obtained from the network-based GBA approach showed that 27.5% of all gene sets had a good classified performance with AUC >0.5. Most significantly, 3 gene sets with AUC >0.6 were denoted as seed gene functions for PDN, including binding, molecular function and regulation of the metabolic process. In summary, we predicted 3 seed gene functions for PDN compared with non-progressors utilizing network-based GBA algorithm. The findings provide insights to reveal pathological and molecular mechanism underlying PDN.

Systems Biology and Machine Learning in Plant-Pathogen Interactions

Systems biology is an inclusive approach to study the static and dynamic emergent properties on a global scale by integrating multiomics datasets to establish qualitative and quantitative associations among multiple biological components. With an abundance of improved high throughput -omics datasets, network-based analyses and machine learning technologies are playing a pivotal role in comprehensive understanding of biological systems. Network topological features reveal most important nodes within a network as well as prioritize significant molecular components for diverse biological networks, including coexpression, protein-protein interaction, and gene regulatory networks. Machine learning techniques provide enormous predictive power through specific feature extraction from biological data. Deep learning, a subtype of machine learning, has plausible future applications because a domain expert for feature extraction is not needed in this algorithm. Inspired by diverse domains of biology, we here review classic systems biology techniques applied in plant immunity thus far. We also discuss additional advanced approaches in both graph theory and machine learning, which may provide new insights for understanding plant-microbe interactions. Finally, we propose a hybrid approach in plant immune systems that harnesses the power of both network biology and machine learning, with a potential to be applicable to both model systems and agronomically important crop plants.

Network-Guided Discovery of Influenza Virus Replication Host Factors

Integrating virus-host interactions with host protein-protein interactions, we have created a method using these established network practices to identify host factors (i.e., proteins) that are likely candidates for antiviral drug targeting. We demonstrate that interaction cascades between host proteins that directly interact with viral proteins and host factors that are important to influenza virus replication are enriched for signaling and immune processes. Additionally, we show that host proteins that interact with viral proteins are in network locations of power. Finally, we demonstrate a new network methodology to predict novel host factors and validate predictions with an siRNA screen. Our results show that integrating virus-host proteins interactions is useful in the identification of antiviral drug target candidates.

The positions of host factors required for viral replication within a human protein-protein interaction (PPI) network can be exploited to identify drug targets that are robust to drug-mediated selective pressure. Host factors can physically interact with viral proteins, be a component of virus-regulated pathways (where proteins do not interact with viral proteins), or be required for viral replication but unregulated by viruses. Here, we demonstrate a method of combining human PPI networks with virus-host PPI data to improve antiviral drug discovery for influenza viruses by identifying target host proteins. Analysis shows that influenza virus proteins physically interact with host proteins in network positions significant for information flow, even after the removal of known abundance-degree bias within PPI data. We have isolated a subnetwork of the human PPI network that connects virus-interacting host proteins to host factors that are important for influenza virus replication without physically interacting with viral proteins. The subnetwork is enriched for signaling and immune processes distinct from those associated with virus-interacting proteins. Selecting proteins based on subnetwork topology, we performed an siRNA screen to determine whether the subnetwork was enriched for virus replication host factors and whether network position within the subnetwork offers an advantage in prioritization of drug targets to control influenza virus replication. We found that the subnetwork is highly enriched for target host proteins—more so than the set of host factors that physically interact with viral proteins. Our findings demonstrate that network positions are a powerful predictor to guide antiviral drug candidate prioritization.

Network‑based gene function inference method to predict optimal gene functions associated with fetal growth restriction

The guilt by association (GBA) principle has been widely used to predict gene functions, and a network‑based approach may enhance the confidence and stability of the analysis compared with focusing on individual genes. Fetal growth restriction (FGR), is the second primary cause of perinatal mortality. Therefore, the present study aimed to predict the optimal gene functions for FGR using a network‑based GBA method. The method was comprised of four parts: Identification of differentially‑expressed genes (DEGs) between patients with FGR and normal controls based on gene expression data; construction of a co‑expression network (CEN) dependent on DEGs, using the Spearman correlation coefficient algorithm; collection of gene ontology (GO) data on the basis of a known confirmed database and DEGs; and prediction of optimal gene functions using the GBA algorithm, for which the area under the receiver operating characteristic curve (AUC) was obtained for each GO term. A total of 115 DEGs and 109 GO terms were obtained for subsequent analysis. All DEGs were mapped to the CEN and formed 6,555 edges. The results of GBA algorithm demonstrated that 78 GO terms had a good classification performance with AUC >0.5. In particular, the AUC for 5 of the GO terms was >0.7, and these were defined as optimal gene functions, including defense response, immune system process, response to stress, cellular response to chemical stimulus and positive regulation of biological process. In conclusion, the results of the present study provided insights into the pathological mechanism underlying FGR, and provided potential biomarkers for early detection and targeted treatment of this disease. However, the interactions between the 5 GO terms remain unclear, and further studies are required.

Disease genes prediction by HMM based PU-learning using gene expression profiles

Predicting disease candidate genes from human genome is a crucial part of nowadays biomedical research. According to observations, diseases with the same phenotype have the similar biological characteristics and genes associated with these same diseases tend to share common functional properties. Therefore, by applying machine learning methods, new disease genes are predicted based on previous ones. In recent studies, some semi-supervised learning methods, called Positive-Unlabeled Learning (PU-Learning) are used for predicting disease candidate genes. In this study, a novel method is introduced to predict disease candidate genes through gene expression profiles by learning hidden Markov models. In order to evaluate the proposed method, it is applied on a mixed part of 398 disease genes from three disease types and 12001 unlabeled genes. Compared to the other methods in literature, the experimental results indicate a significant improvement in favor of the proposed method.

Unveiling network-based functional features through integration of gene expression into protein networks

Decoding health and disease phenotypes is one of the fundamental objectives in biomedicine. Whereas high-throughput omics approaches are available, it is evident that any single omics approach might not be adequate to capture the complexity of phenotypes. Therefore, integrated multi-omics approaches have been used to unravel genotype-phenotype relationships such as global regulatory mechanisms and complex metabolic networks in different eukaryotic organisms. Some of the progress and challenges associated with integrated omics studies have been reviewed previously in comprehensive studies. In this work, we highlight and review the progress, challenges and advantages associated with emerging approaches, integrating gene expression and protein-protein interaction networks to unravel network-based functional features. This includes identifying disease related genes, gene prioritization, clustering protein interactions, developing the modules, extract active subnetworks and static protein complexes or dynamic/temporal protein complexes. We also discuss how these approaches contribute to our understanding of the biology of complex traits and diseases. This article is part of a Special Issue entitled: Cardiac adaptations to obesity, diabetes and insulin resistance, edited by Professors Jan F.C. Glatz, Jason R.B. Dyck and Christine Des Rosiers.

Integrated regulatory network reveals novel candidate regulators in the development of negative energy balance in cattle

Negative energy balance (NEB) is an altered metabolic state in modern high-yielding dairy cows. This metabolic state occurs in the early postpartum period when energy demands for milk production and maintenance exceed that of energy intake. Negative energy balance or poor adaptation to this metabolic state has important effects on the liver and can lead to metabolic disorders and reduced fertility. The roles of regulatory factors, including transcription factors (TFs) and micro RNAs (miRNAs) have often been separately studied for evaluating of NEB. However, adaptive response to NEB is controlled by complex gene networks and still not fully understood. In this study, we aimed to discover the integrated gene regulatory networks involved in NEB development in liver tissue. We downloaded data sets including mRNA and miRNA expression profiles related to three and four cows with severe and moderate NEB, respectively. Our method integrated two independent types of information: module inference network by TFs, miRNAs and mRNA expression profiles (RNA-seq data) and computational target predictions. In total, 176 modules were predicted by using gene expression data and 64 miRNAs and 63 TFs were assigned to these modules. By using our integrated computational approach, we identified 13 TF-module and 19 miRNA-module interactions. Most of these modules were associated with liver metabolic processes as well as immune and stress responses, which might play crucial roles in NEB development. Literature survey results also showed that several regulators and gene targets have already been characterized as important factors in liver metabolic processes. These results provided novel insights into regulatory mechanisms at the TF and miRNA levels during NEB. In addition, the method described in this study seems to be applicable to construct integrated regulatory networks for different diseases or disorders.

Gene gravity-like algorithm for disease gene prediction based on phenotype-specific network

Background

Polygenic diseases are usually caused by the dysfunction of multiple genes. Unravelling such disease genes is crucial to fully understand the genetic landscape of diseases on molecular level. With the advent of ‘omic’ data era, network-based methods have prominently boosted disease gene discovery. However, how to make better use of different types of data for the prediction of disease genes remains a challenge.

Results

In this study, we improved the performance of disease gene prediction by integrating the similarity of disease phenotype, biological function and network topology. First, for each phenotype, a phenotype-specific network was specially constructed by mapping phenotype similarity information of given phenotype onto the protein-protein interaction (PPI) network. Then, we developed a gene gravity-like algorithm, to score candidate genes based on not only topological similarity but also functional similarity. We tested the proposed network and algorithm by conducting leave-one-out and leave-10%-out cross validation and compared them with state-of-art algorithms. The results showed a preference to phenotype-specific network as well as gene gravity-like algorithm. At last, we tested the predicting capacity of proposed algorithms by test gene set derived from the DisGeNET database. Also, potential disease genes of three polygenic diseases, obesity, prostate cancer and lung cancer, were predicted by proposed methods. We found that the predicted disease genes are highly consistent with literature and database evidence.

Conclusions

The good performance of phenotype-specific networks indicates that phenotype similarity information has positive effect on the prediction of disease genes. The proposed gene gravity-like algorithm outperforms the algorithm of Random Walk with Restart (RWR), implicating its predicting capacity by combing topological similarity with functional similarity. Our work will give an insight to the discovery of disease genes by fusing multiple similarities of genes and diseases.

Electronic supplementary material

The online version of this article (10.1186/s12918-017-0519-9) contains supplementary material, which is available to authorized users.

Significance and Functional Similarity for Identification of Disease Genes

One of the most significant research issues in functional genomics is insilico identification of disease related genes. In this regard, the paper presents a new gene selection algorithm, termed as SiFS, for identification of disease genes. It integrates the information obtained from interaction network of proteins and gene expression profiles. The proposed SiFS algorithm culls out a subset of genes from microarray data as disease genes by maximizing both significance and functional similarity of the selected gene subset. Based on the gene expression profiles, the significance of a gene with respect to another gene is computed using mutual information. On the other hand, a new measure of similarity is introduced to compute the functional similarity between two genes. Information derived from the protein-protein interaction network forms the basis of the proposed SiFS algorithm. The performance of the proposed gene selection algorithm and new similarity measure, is compared with that of other related methods and similarity measures, using several cancer microarray data sets.

Predicting pathogenic genes for primary myelofibrosis based on a system‑network approach

The aim of the present study was to predict pathogenic genes for primary myelofibrosis (PMF) using a system‑network approach by combining protein‑protein interaction (PPI) network and gene expression data with known pathogenic genes. PMF gene expression profiles, known pathogenic genes and protein‑protein interactions were obtained. Using these data, differentially expressed genes (DEGs) were identified between PMF and normal conditions using significance analysis of microarrays, and seed genes were determined based on the intersection of known pathogenic genes and the PMF gene expression profile. A new network was constructed using the seed genes and their adjacent DEGs within the PPI network. Subsequently, a pathogenic network was extracted from the new network, and contained genes that interacted with at least two seed genes, and the candidate pathogenic genes were predicted based on the cohesion with seed genes. Cluster analysis was performed to mine the pathogenic modules from the pathogenic network, and functional analysis was performed to identify the putative biological processes of the candidate pathogenic genes. Results from the present study identified 845 DEGs between PMF and normal conditions, and 45 seed genes in PMF were screened. Subsequently, a pathogenic network comprising 103 nodes and 265 interactions was constructed, and 4 pathogenic modules (modules A‑D) were mined from the pathogenic network. There were nine candidate pathogenic genes contained within Module A and four potential pathogenic genes, including E1A‑binding protein p300, RAS‑like proto‑oncogene A, von Willebrand factor and RAF‑1 proto‑oncogene, serine/threonine kinase, were identified that may be involved in the same biological process with the seed genes. This study predicted 10 candidate pathogenic genes and several signaling pathways that may be related to the pathogenesis of PMF using a system‑network approach. These predictions may shed light on the PMF pathogenesis and may provide guidelines for future experimental verification.

Fusion of expression values and protein interaction information using multi-objective optimization for improving gene clustering

One of the crucial problems in the field of functional genomics is to identify a set of genes which are responsible for a particular cellular mechanism. The current work explores the usage of a multi-objective optimization based genetic clustering technique to classify genes into groups with respect to their functional similarities and biological relevance. Our contribution is two-fold: firstly a new quality measure to compute the goodness of gene-clusters namely protein-protein interaction confidence score is developed. This utilizes the confidence scores of the protein-protein interaction networks to measure the similarity between genes of a particular cluster with respect to their biochemical protein products. Secondly, a multi-objective based clustering approach is developed which intelligently uses integrated information of expression values of microarray dataset and protein-protein interaction confidence scores to select both statistically and biologically relevant genes. For that very purpose, some biological cluster validity indices, viz. biological homogeneity index and protein-protein interaction confidence score, along with two traditional internal cluster validity indices, viz. fuzzy partition coefficient and Pakhira-Bandyopadhyay-Maulik-index, are simultaneously optimized during the clustering process. Experimental results on three real-life gene expression datasets show that the addition of new objective capturing protein-protein interaction information aids in clustering the genes as compared to the existing techniques. The observations are further supported by biological and statistical significance tests.

Identifying key genes in glaucoma based on a benchmarked dataset and the gene regulatory network

The current study aimed to identify key genes in glaucoma based on a benchmarked dataset and gene regulatory network (GRN). Local and global noise was added to the gene expression dataset to produce a benchmarked dataset. Differentially-expressed genes (DEGs) between patients with glaucoma and normal controls were identified utilizing the Linear Models for Microarray Data (Limma) package based on benchmarked dataset. A total of 5 GRN inference methods, including Zscore, GeneNet, context likelihood of relatedness (CLR) algorithm, Partial Correlation coefficient with Information Theory (PCIT) and GEne Network Inference with Ensemble of Trees (Genie3) were evaluated using receiver operating characteristic (ROC) and precision and recall (PR) curves. The interference method with the best performance was selected to construct the GRN. Subsequently, topological centrality (degree, closeness and betweenness) was conducted to identify key genes in the GRN of glaucoma. Finally, the key genes were validated by performing reverse transcription-quantitative polymerase chain reaction (RT-qPCR). A total of 176 DEGs were detected from the benchmarked dataset. The ROC and PR curves of the 5 methods were analyzed and it was determined that Genie3 had a clear advantage over the other methods; thus, Genie3 was used to construct the GRN. Following topological centrality analysis, 14 key genes for glaucoma were identified, including IL6, EPHA2 and GSTT1 and 5 of these 14 key genes were validated by RT-qPCR. Therefore, the current study identified 14 key genes in glaucoma, which may be potential biomarkers to use in the diagnosis of glaucoma and aid in identifying the molecular mechanism of this disease.

HAPPI-2: a Comprehensive and High-quality Map of Human Annotated and Predicted Protein Interactions

BACKGROUND

Human protein-protein interaction (PPI) data is essential to network and systems biology studies. PPI data can help biochemists hypothesize how proteins form complexes by binding to each other, how extracellular signals propagate through post-translational modification of de-activated signaling molecules, and how chemical reactions are coupled by enzymes involved in a complex biological process. Our capability to develop good public database resources for human PPI data has a direct impact on the quality of future research on genome biology and medicine.

RESULTS

The database of Human Annotated and Predicted Protein Interactions (HAPPI) version 2.0 is a major update to the original HAPPI 1.0 database. It contains 2,922,202 unique protein-protein interactions (PPI) linked by 23,060 human proteins, making it the most comprehensive database covering human PPI data today. These PPIs contain both physical/direct interactions and high-quality functional/indirect interactions. Compared with the HAPPI 1.0 database release, HAPPI database version 2.0 (HAPPI-2) represents a 485% of human PPI data coverage increase and a 73% protein coverage increase. The revamped HAPPI web portal provides users with a friendly search, curation, and data retrieval interface, allowing them to retrieve human PPIs and available annotation information on the interaction type, interaction quality, interacting partner drug targeting data, and disease information. The updated HAPPI-2 can be freely accessed by Academic users at http://discovery.informatics.uab.edu/HAPPI .

CONCLUSIONS

While the underlying data for HAPPI-2 are integrated from a diverse data sources, the new HAPPI-2 release represents a good balance between data coverage and data quality of human PPIs, making it ideally suited for network biology.

The integration of weighted human gene association networks based on link prediction

Background

Physical and functional interplays between genes or proteins have important biological meaning for cellular functions. Some efforts have been made to construct weighted gene association meta-networks by integrating multiple biological resources, where the weight indicates the confidence of the interaction. However, it is found that these existing human gene association networks share only quite limited overlapped interactions, suggesting their incompleteness and noise.

Results

Here we proposed a workflow to construct a weighted human gene association network using information of six existing networks, including two weighted specific PPI networks and four gene association meta-networks. We applied link prediction algorithm to predict possible missing links of the networks, cross-validation approach to refine each network and finally integrated the refined networks to get the final integrated network.

Conclusions

The common information among the refined networks increases notably, suggesting their higher reliability. Our final integrated network owns much more links than most of the original networks, meanwhile its links still keep high functional relevance. Being used as background network in a case study of disease gene prediction, the final integrated network presents good performance, implying its reliability and application significance. Our workflow could be insightful for integrating and refining existing gene association data.

Electronic supplementary material

The online version of this article (doi:10.1186/s12918-017-0398-0) contains supplementary material, which is available to authorized users.

An integrative approach to investigate the respective roles of single-nucleotide variants and copy-number variants in Attention-Deficit/Hyperactivity Disorder

Many studies have attempted to investigate the genetic susceptibility of Attention-Deficit/Hyperactivity Disorder (ADHD), but without much success. The present study aimed to analyze both single-nucleotide and copy-number variants contributing to the genetic architecture of ADHD. We generated exome data from 30 Brazilian trios with sporadic ADHD. We also analyzed a Brazilian sample of 503 children/adolescent controls from a High Risk Cohort Study for the Development of Childhood Psychiatric Disorders, and also previously published results of five CNV studies and one GWAS meta-analysis of ADHD involving children/adolescents. The results from the Brazilian trios showed that cases with de novo SNVs tend not to have de novo CNVs and vice-versa. Although the sample size is small, we could also see that various comorbidities are more frequent in cases with only inherited variants. Moreover, using only genes expressed in brain, we constructed two "in silico" protein-protein interaction networks, one with genes from any analysis, and other with genes with hits in two analyses. Topological and functional analyses of genes in this network uncovered genes related to synapse, cell adhesion, glutamatergic and serotoninergic pathways, both confirming findings of previous studies and capturing new genes and genetic variants in these pathways.

Systematic interaction network filtering identifies CRMP1 as a novel suppressor of huntingtin misfolding and neurotoxicity

Assemblies of huntingtin (HTT) fragments with expanded polyglutamine (polyQ) tracts are a pathological hallmark of Huntington's disease (HD). The molecular mechanisms by which these structures are formed and cause neuronal dysfunction and toxicity are poorly understood. Here, we utilized available gene expression data sets of selected brain regions of HD patients and controls for systematic interaction network filtering in order to predict disease-relevant, brain region-specific HTT interaction partners. Starting from a large protein-protein interaction (PPI) data set, a step-by-step computational filtering strategy facilitated the generation of a focused PPI network that directly or indirectly connects 13 proteins potentially dysregulated in HD with the disease protein HTT. This network enabled the discovery of the neuron-specific protein CRMP1 that targets aggregation-prone, N-terminal HTT fragments and suppresses their spontaneous self-assembly into proteotoxic structures in various models of HD. Experimental validation indicates that our network filtering procedure provides a simple but powerful strategy to identify disease-relevant proteins that influence misfolding and aggregation of polyQ disease proteins.

Screening key genes associated with congenital heart defects in Down syndrome based on differential expression network

BACKGROUND

Down syndrome (DS) is the most common viable chromosomal disorder with intellectual impairment and several other developmental abnormalities. Forty to fifty percent of newborns with DS have some form of congenital heart defects (CHD). The genome of CHD in DS has already been obtained, but the underlying genomic or gene expression variation that contributes to the manifestation of a CHD in DS is still unknown.

OBJECTIVE

This study was aimed to analyze key genes of patients with CHD in DS.

METHODS

Differential expression network (DEN) approach was employed to analyze the dyeregulated genes and pathways in this study. First, the differentially expressed genes (DEGs) between CHD in DS and normal subjects were screened based on the microarray expression data. Next, the differential interactions were identified using spearman correlation coefficients of edges in different conditions. The DEN was then constructed combining both DEGs and differential interactions, and HUB genes were gained by degree centrality analysis of DEN. Meanwhile, disease genes included in the DEN were also ascertained.

RESULTS

When analyzing gene expression values in different conditions, no DEGs were identified. While, a total of 984 gene pairs with significant differential expression were identified. Finally, the DEN was constructed only using differential edges in our study. In this network, eight HUB genes were identified, and thereinto four genes (UBC, APP, HUWE1 and SRC) were both HUB genes and disease genes.

CONCLUSIONS

DEN approach should be taken as a useful complement to traditional differential genes methods. We provide several potential underlying biomarkers for CHD in DS.

Identification of active miRNA and transcription factor regulatory pathways in human obesity-related inflammation

Background

Obesity-induced chronic inflammation plays a fundamental role in the pathogenesis of metabolic syndrome (MS). Recently, a growing body of evidence supports that miRNAs are largely dysregulated in obesity and that specific miRNAs regulate obesity-associated inflammation. We applied an approach aiming to identify active miRNA-TF-gene regulatory pathways in obesity. Firstly, we detected differentially expressed genes (DEGs) and differentially expressed miRNAs (DEmiRs) from mRNA and miRNA expression profiles, respectively. Secondly, by mapping the DEGs and DEmiRs to the curated miRNA-TF-gene regulatory network as active seed nodes and connect them with their immediate neighbors, we obtained the potential active miRNA-TF-gene regulatory subnetwork in obesity. Thirdly, using a Breadth-First-Search (BFS) algorithm, we identified potential active miRNA-TF-gene regulatory pathways in obesity. Finally, through the hypergeometric test, we identified the active miRNA-TF-gene regulatory pathways that were significantly related to obesity.

Results

The potential active pathways with FDR < 0.0005 were considered to be the active miRNA-TF regulatory pathways in obesity. The union of the active pathways is visualized and identical nodes of the active pathways were merged.

Conclusions

We identified 23 active miRNA-TF-gene regulatory pathways that were significantly related to obesity-related inflammation.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-015-0512-5) contains supplementary material, which is available to authorized users.

GroupRank: rank candidate genes in PPI network by differentially expressed gene groups

Many cell activities are organized as a network, and genes are clustered into co-expressed groups if they have the same or closely related biological function or they are co-regulated. In this study, based on an assumption that a strong candidate disease gene is more likely close to gene groups in which all members coordinately differentially express than individual genes with differential expression, we developed a novel disease gene prioritization method GroupRank by integrating gene co-expression and differential expression information generated from microarray data as well as PPI network. A candidate gene is ranked high using GroupRank if it is differentially expressed in disease and control or is close to differentially co-expressed groups in PPI network. We tested our method on data sets of lung, kidney, leukemia and breast cancer. The results revealed GroupRank could efficiently prioritize disease genes with significantly improved AUC value in comparison to the previous method with no consideration of co-exprssed gene groups in PPI network. Moreover, the functional analyses of the major contributing gene group in gene prioritization of kidney cancer verified that our algorithm GroupRank not only ranks disease genes efficiently but also could help us identify and understand possible mechanisms in important physiological and pathological processes of disease.

The network organization of cancer-associated protein complexes in human tissues

Differential gene expression profiles for detecting disease genes have been studied intensively in systems biology. However, it is known that various biological functions achieved by proteins follow from the ability of the protein to form complexes by physically binding to each other. In other words, the functional units are often protein complexes rather than individual proteins. Thus, we seek to replace the perspective of disease-related genes by disease-related complexes, exemplifying with data on 39 human solid tissue cancers and their original normal tissues. To obtain the differential abundance levels of protein complexes, we apply an optimization algorithm to genome-wide differential expression data. From the differential abundance of complexes, we extract tissue- and cancer-selective complexes, and investigate their relevance to cancer. The method is supported by a clustering tendency of bipartite cancer-complex relationships, as well as a more concrete and realistic approach to disease-related proteomics.

Spatio-temporal analysis of type 2 diabetes mellitus based on differential expression networks

T2DM is complex in its dynamical dependence on multiple tissues, disease states, and factors' interactions. However, most existing work devoted to characterizing its pathophysiology from one static tissue, individual factors, or single state. Here we perform a spatio-temporal analysis on T2DM by developing a new form of molecular network, i.e. 'differential expression network' (DEN), which can reflect phenotype differences at network level. Static DENs show that three tissues (white adipose, skeletal muscle, and liver) all suffer from severe inflammation and perturbed metabolism, among which metabolic functions are seriously affected in liver. Dynamical analysis on DENs reveals metabolic function changes in adipose and liver are consistent with insulin resistance (IR) deterioration. Close investigation on IR pathway identifies 'disease interactions', revealing that IR deterioration is earlier than that on SlC2A4 in adipose and muscle. Our analysis also provides evidence that rising of insulin secretion is the root cause of IR in diabetes.

Structure and dynamics of molecular networks: a novel paradigm of drug discovery: a comprehensive review

Despite considerable progress in genome- and proteome-based high-throughput screening methods and in rational drug design, the increase in approved drugs in the past decade did not match the increase of drug development costs. Network description and analysis not only give a systems-level understanding of drug action and disease complexity, but can also help to improve the efficiency of drug design. We give a comprehensive assessment of the analytical tools of network topology and dynamics. The state-of-the-art use of chemical similarity, protein structure, protein-protein interaction, signaling, genetic interaction and metabolic networks in the discovery of drug targets is summarized. We propose that network targeting follows two basic strategies. The "central hit strategy" selectively targets central nodes/edges of the flexible networks of infectious agents or cancer cells to kill them. The "network influence strategy" works against other diseases, where an efficient reconfiguration of rigid networks needs to be achieved by targeting the neighbors of central nodes/edges. It is shown how network techniques can help in the identification of single-target, edgetic, multi-target and allo-network drug target candidates. We review the recent boom in network methods helping hit identification, lead selection optimizing drug efficacy, as well as minimizing side-effects and drug toxicity. Successful network-based drug development strategies are shown through the examples of infections, cancer, metabolic diseases, neurodegenerative diseases and aging. Summarizing >1200 references we suggest an optimized protocol of network-aided drug development, and provide a list of systems-level hallmarks of drug quality. Finally, we highlight network-related drug development trends helping to achieve these hallmarks by a cohesive, global approach.

Therapeutic effects of astragaloside IV on myocardial injuries: multi-target identification and network analysis

Astragaloside IV (AGS-IV) is a main active ingredient of Astragalus membranaceus Bunge, a medicinal herb used for cardiovascular diseases (CVD). In this work, we investigated the therapeutic mechanisms of AGS-IV at a network level by computer-assisted target identification with the in silico inverse docking program (INVDOCK). Targets included in the analysis covered all signaling pathways thought to be implicated in the therapeutic actions of all CVD drugs approved by US FDA. A total of 39 putative targets were identified. Three of these targets, calcineurin (CN), angiotensin-converting enzyme (ACE), and c-Jun N-terminal kinase (JNK), were experimentally validated at a molecular level. Protective effects of AGS-IV were also compared with the CN inhibitor cyclosporin A (CsA) in cultured cardiomyocytes exposed to adriamycin. Network analysis of protein-protein interactions (PPI) was carried out with reference to the therapeutic profiles of approved CVD drugs. The results suggested that the therapeutic effects of AGS-IV are based upon a combination of blocking calcium influx, vasodilation, anti-thrombosis, anti-oxidation, anti-inflammation and immune regulation.

Insights into the pathogenesis of axial spondyloarthropathy from network and pathway analysis

Background

Complex chronic diseases are usually not caused by changes in a single causal gene but by an unbalanced regulating network resulting from the dysfunctions of multiple genes or their products. Therefore, network based systems approach can be helpful for the identification of candidate genes related to complex diseases and their relationships. Axial spondyloarthropathy (SpA) is a group of chronic inflammatory joint diseases that mainly affect the spine and the sacroiliac joints. The pathogenesis of SpA remains largely unknown.

Results

In this paper, we conducted a network study of the pathogenesis of SpA. We integrated data related to SpA, from the OMIM database, proteomics and microarray experiments of SpA, to prioritize SpA candidate disease genes in the context of human protein interactome. Based on the top ranked SpA related genes, we constructed a SpA specific PPI network, identified potential pathways associated with SpA, and finally sketched an overview of biological processes involved in the development of SpA.

Conclusions

The protein-protein interaction (PPI) network and pathways reflect the link between the two pathological processes of SpA, i.e., immune mediated inflammation, as well as imbalanced bone modelling caused new boneformation and bone loss. We found that some known disease causative genes, such as TNFand ILs, play pivotal roles in this interaction.