Predicting disease genes using protein-protein interactions

A fast and high performance multiple data integration algorithm for identifying human disease genes

BACKGROUND

Integrating multiple data sources is indispensable in improving disease gene identification. It is not only due to the fact that disease genes associated with similar genetic diseases tend to lie close with each other in various biological networks, but also due to the fact that gene-disease associations are complex. Although various algorithms have been proposed to identify disease genes, their prediction performances and the computational time still should be further improved.

RESULTS

In this study, we propose a fast and high performance multiple data integration algorithm for identifying human disease genes. A posterior probability of each candidate gene associated with individual diseases is calculated by using a Bayesian analysis method and a binary logistic regression model. Two prior probability estimation strategies and two feature vector construction methods are developed to test the performance of the proposed algorithm.

CONCLUSIONS

The proposed algorithm is not only generated predictions with high AUC scores, but also runs very fast. When only a single PPI network is employed, the AUC score is 0.769 by using F2 as feature vectors. The average running time for each leave-one-out experiment is only around 1.5 seconds. When three biological networks are integrated, the AUC score using F3 as feature vectors increases to 0.830, and the average running time for each leave-one-out experiment takes only about 12.54 seconds. It is better than many existing algorithms.

ProSim: A Method for Prioritizing Disease Genes Based on Protein Proximity and Disease Similarity

Predicting disease genes for a particular genetic disease is very challenging in bioinformatics. Based on current research studies, this challenge can be tackled via network-based approaches. Furthermore, it has been highlighted that it is necessary to consider disease similarity along with the protein's proximity to disease genes in a protein-protein interaction (PPI) network in order to improve the accuracy of disease gene prioritization. In this study we propose a new algorithm called proximity disease similarity algorithm (ProSim), which takes both of the aforementioned properties into consideration, to prioritize disease genes. To illustrate the proposed algorithm, we have conducted six case studies, namely, prostate cancer, Alzheimer's disease, diabetes mellitus type 2, breast cancer, colorectal cancer, and lung cancer. We employed leave-one-out cross validation, mean enrichment, tenfold cross validation, and ROC curves to evaluate our proposed method and other existing methods. The results show that our proposed method outperforms existing methods such as PRINCE, RWR, and DADA.

Estimating a ranked list of human hereditary diseases for clinical phenotypes by using weighted bipartite network

With the availability of the huge medical knowledge data on the Internet such as the human disease network, protein-protein interaction (PPI) network, and phenotypegene, gene-disease bipartite networks, it becomes practical to help doctors by suggesting plausible hereditary diseases for a set of clinical phenotypes. However, identifying candidate diseases that best explain a set of clinical phenotypes by considering various heterogeneous networks is still a challenging task. In this paper, we propose a new method for estimating a ranked list of plausible diseases by associating phenotypegene with gene-disease bipartite networks. Our approach is to count the frequency of all the paths from a phenotype to a disease through their associated causative genes, and link the phenotype to the disease with path frequency in a new phenotype-disease bipartite (PDB) network. After that, we generate the candidate weights for the edges of phenotypes with diseases in PDB network. We evaluate our proposed method in terms of Normalized Discounted Cumulative Gain (NDCG), and demonstrate that we outperform the previously known disease ranking method called Phenomizer.

Dynamics of the discovery process of protein-protein interactions from low content studies

Background

Thousands of biological and biomedical investigators study of the functional role of single genes and their protein products in normal physiology and in disease. The findings from these studies are reported in research articles that stimulate new research. It is now established that a complex regulatory networks's is controlling human cellular fate, and this community of researchers are continually unraveling this network topology. Attempts to integrate results from such accumulated knowledge resulted in literature-based protein-protein interaction networks (PPINs) and pathway databases. These databases are widely used by the community to analyze new data collected from emerging genome-wide studies with the assumption that the data within these literature-based databases is the ground truth and contain no biases. While suspicion for research focus biases is growing, a concrete proof for it is still missing. It is difficult to prove because the real PPINs are mostly unknown.

Results

Here we analyzed the longitudinal discovery process of literature-based mammalian and yeast PPINs to observe that these networks are discovered non-uniformly. The pattern of discovery is related to a theoretical concept proposed by Kauffman called “expanding the adjacent possible”. We introduce a network discovery model which explicitly includes the space of possibilities in the form of a true underlying PPIN.

Conclusions

Our model strongly suggests that research focus biases exist in the observed discovery dynamics of these networks. In summary, more care should be placed when using PPIN databases for analysis of newly acquired data, and when considering prior knowledge when designing new experiments.

Electronic supplementary material

The online version of this article (doi:10.1186/s12918-015-0173-z) contains supplementary material, which is available to authorized users.

Common variants in the MKL1 gene confer risk of schizophrenia

Genome-wide association studies (GWAS) of schizophrenia have identified multiple risk variants with robust association signals for schizophrenia. However, these variants could explain only a small proportion of schizophrenia heritability. Furthermore, the effect size of these risk variants is relatively small (eg, most of them had an OR less than 1.2), suggesting that additional risk variants may be detected when increasing sample size in analysis. Here, we report the identification of a genome-wide significant schizophrenia risk locus at 22q13.1 by combining 2 large-scale schizophrenia cohort studies. Our meta-analysis revealed that 7 single nucleotide polymorphism (SNPs) on chromosome 22q13.1 reached the genome-wide significance level (P < 5.0×10(-8)) in the combined samples (a total of 38441 individuals). Among them, SNP rs6001946 had the most significant association with schizophrenia (P = 2.04×10(-8)). Interestingly, all 7 SNPs are in high linkage disequilibrium and located in the MKL1 gene. Expression analysis showed that MKL1 is highly expressed in human and mouse brains. We further investigated functional links between MKL1 and proteins encoded by other schizophrenia susceptibility genes in the whole human protein interaction network. We found that MKL1 physically interacts with GSK3B, a protein encoded by a well-characterized schizophrenia susceptibility gene. Collectively, our results revealed that genetic variants in MKL1 might confer risk to schizophrenia. Further investigation of the roles of MKL1 in the pathogenesis of schizophrenia is warranted.

A DIseAse MOdule Detection (DIAMOnD) algorithm derived from a systematic analysis of connectivity patterns of disease proteins in the human interactome

The observation that disease associated proteins often interact with each other has fueled the development of network-based approaches to elucidate the molecular mechanisms of human disease. Such approaches build on the assumption that protein interaction networks can be viewed as maps in which diseases can be identified with localized perturbation within a certain neighborhood. The identification of these neighborhoods, or disease modules, is therefore a prerequisite of a detailed investigation of a particular pathophenotype. While numerous heuristic methods exist that successfully pinpoint disease associated modules, the basic underlying connectivity patterns remain largely unexplored. In this work we aim to fill this gap by analyzing the network properties of a comprehensive corpus of 70 complex diseases. We find that disease associated proteins do not reside within locally dense communities and instead identify connectivity significance as the most predictive quantity. This quantity inspires the design of a novel Disease Module Detection (DIAMOnD) algorithm to identify the full disease module around a set of known disease proteins. We study the performance of the algorithm using well-controlled synthetic data and systematically validate the identified neighborhoods for a large corpus of diseases.

Diseases are rarely the result of an abnormality in a single gene, but involve a whole cascade of interactions between several cellular processes. To disentangle these complex interactions it is necessary to study genotype-phenotype relationships in the context of protein-protein interaction networks. Our analysis of 70 diseases shows that disease proteins are not randomly scattered within these networks, but agglomerate in specific regions, suggesting the existence of specific disease modules for each disease. The identification of these modules is the first step towards elucidating the biological mechanisms of a disease or for a targeted search of drug targets. We present a systematic analysis of the connectivity patterns of disease proteins and determine the most predictive topological property for their identification. This allows us to rationally design a reliable and efficient Disease Module Detection algorithm (DIAMOnD).

Integrative analysis of kinase networks in TRAIL-induced apoptosis provides a source of potential targets for combination therapy

Tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) is an endogenous secreted peptide and, in preclinical studies, preferentially induces apoptosis in tumor cells rather than in normal cells. The acquisition of resistance in cells exposed to TRAIL or its mimics limits their clinical efficacy. Because kinases are intimately involved in the regulation of apoptosis, we systematically characterized kinases involved in TRAIL signaling. Using RNA interference (RNAi) loss-of-function and cDNA overexpression screens, we identified 169 protein kinases that influenced the dynamics of TRAIL-induced apoptosis in the colon adenocarcinoma cell line DLD-1. We classified the kinases as sensitizers or resistors or modulators, depending on the effect that knockdown and overexpression had on TRAIL-induced apoptosis. Two of these kinases that were classified as resistors were PX domain-containing serine/threonine kinase (PXK) and AP2-associated kinase 1 (AAK1), which promote receptor endocytosis and may enable cells to resist TRAIL-induced apoptosis by enhancing endocytosis of the TRAIL receptors. We assembled protein interaction maps using mass spectrometry-based protein interaction analysis and quantitative phosphoproteomics. With these protein interaction maps, we modeled information flow through the networks and identified apoptosis-modifying kinases that are highly connected to regulated substrates downstream of TRAIL. The results of this analysis provide a resource of potential targets for the development of TRAIL combination therapies to selectively kill cancer cells.

Mining disease genes using integrated protein-protein interaction and gene-gene co-regulation information

•

An eQTL-based gene–gene co-regulation network was constructed.

•

We adopted a random walk with restart (RWR) algorithm to mine for Alzheimer-disease related genes.

•

The integrated HPRD PPI and GGCRN network had faster convergence than using HPRD PPI alone.

•

The integrated network also revealed new disease-related genes.

In humans, despite the rapid increase in disease-associated gene discovery, a large proportion of disease-associated genes are still unknown. Many network-based approaches have been used to prioritize disease genes. Many networks, such as the protein–protein interaction (PPI), KEGG, and gene co-expression networks, have been used. Expression quantitative trait loci (eQTLs) have been successfully applied for the determination of genes associated with several diseases. In this study, we constructed an eQTL-based gene–gene co-regulation network (GGCRN) and used it to mine for disease genes. We adopted the random walk with restart (RWR) algorithm to mine for genes associated with Alzheimer disease. Compared to the Human Protein Reference Database (HPRD) PPI network alone, the integrated HPRD PPI and GGCRN networks provided faster convergence and revealed new disease-related genes. Therefore, using the RWR algorithm for integrated PPI and GGCRN is an effective method for disease-associated gene mining.

Prediction of cancer proteins by integrating protein interaction, domain frequency, and domain interaction data using machine learning algorithms

Many proteins are known to be associated with cancer diseases. It is quite often that their precise functional role in disease pathogenesis remains unclear. A strategy to gain a better understanding of the function of these proteins is to make use of a combination of different aspects of proteomics data types. In this study, we extended Aragues's method by employing the protein-protein interaction (PPI) data, domain-domain interaction (DDI) data, weighted domain frequency score (DFS), and cancer linker degree (CLD) data to predict cancer proteins. Performances were benchmarked based on three kinds of experiments as follows: (I) using individual algorithm, (II) combining algorithms, and (III) combining the same classification types of algorithms. When compared with Aragues's method, our proposed methods, that is, machine learning algorithm and voting with the majority, are significantly superior in all seven performance measures. We demonstrated the accuracy of the proposed method on two independent datasets. The best algorithm can achieve a hit ratio of 89.4% and 72.8% for lung cancer dataset and lung cancer microarray study, respectively. It is anticipated that the current research could help understand disease mechanisms and diagnosis.

Predicted protein-protein interactions in the moss Physcomitrella patens: a new bioinformatic resource

Background

Physcomitrella patens, a haploid dominant plant, is fast becoming a useful molecular genetics and bioinformatics tool due to its key phylogenetic position as a bryophyte in the post-genomic era. Genome sequences from select reference species were compared bioinformatically to Physcomitrella patens using reciprocal blasts with the InParanoid software package. A reference protein interaction database assembled using MySQL by compiling BioGrid, BIND, DIP, and Intact databases was queried for moss orthologs existing for both interacting partners. This method has been used to successfully predict interactions for a number of angiosperm plants.

Results

The first predicted protein-protein interactome for a bryophyte based on the interolog method contains 67,740 unique interactions from 5,695 different Physcomitrella patens proteins. Most conserved interactions among proteins were those associated with metabolic processes. Over-represented Gene Ontology categories are reported here.

Conclusion

Addition of moss, a plant representative 200 million years diverged from angiosperms to interactomic research greatly expands the possibility of conducting comparative analyses giving tremendous insight into network evolution of land plants. This work helps demonstrate the utility of “guilt-by-association” models for predicting protein interactions, providing provisional roadmaps that can be explored using experimental approaches. Included with this dataset is a method for characterizing subnetworks and investigating specific processes, such as the Calvin-Benson-Bassham cycle.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-015-0524-1) contains supplementary material, which is available to authorized users.

A t(3;9)(q25.1;q34.3) translocation leading to OLFM1 fusion transcripts in Gilles de la Tourette syndrome, OCD and ADHD

Gilles de la Tourette syndrome (GTS) is a neuropsychiatric disorder with a strong genetic etiology; however, finding of candidate genes is hampered by its genetic heterogeneity and the influence of non-genetic factors on disease pathogenesis. We report a case of a male patient with GTS, obsessive compulsive disorder, attention-deficit/hyperactivity-disorder, as well as other comorbidities, and a translocation t(3;9)(q25.1;q34.3) inherited from a mother with tics. Mate-pair sequencing revealed that the translocation breakpoints truncated the olfactomedin 1 (OLFM1) gene and two uncharacterized transcripts. Reverse-transcription PCR identified several fusion transcripts in the carriers, and OLFM1 expression was found to be high in GTS-related human brain regions. As OLFM1 plays a role in neuronal development it is a likely candidate gene for neuropsychiatric disorders and haploinsufficiency of OLFM1 could be a contributing risk factor to the phenotype of the carriers. In addition, one of the fusion transcripts may exert a dominant-negative or gain-of-function effect. OLFM1 is unlikely to be a major GTS susceptibility gene as no point mutations or copy number variants affecting OLFM1 were identified in 175 additional patients. The translocation described is thus a unique event, but further studies in larger cohorts are required to elucidate involvement of OLFM1 in GTS pathogenesis.

Disease networks. Uncovering disease-disease relationships through the incomplete interactome

According to the disease module hypothesis, the cellular components associated with a disease segregate in the same neighborhood of the human interactome, the map of biologically relevant molecular interactions. Yet, given the incompleteness of the interactome and the limited knowledge of disease-associated genes, it is not obvious if the available data have sufficient coverage to map out modules associated with each disease. Here we derive mathematical conditions for the identifiability of disease modules and show that the network-based location of each disease module determines its pathobiological relationship to other diseases. For example, diseases with overlapping network modules show significant coexpression patterns, symptom similarity, and comorbidity, whereas diseases residing in separated network neighborhoods are phenotypically distinct. These tools represent an interactome-based platform to predict molecular commonalities between phenotypically related diseases, even if they do not share primary disease genes.

Network propagation with dual flow for gene prioritization

Based on the hypothesis that the neighbors of disease genes trend to cause similar diseases, network-based methods for disease prediction have received increasing attention. Taking full advantage of network structure, the performance of global distance measurements is generally superior to local distance measurements. However, some problems exist in the global distance measurements. For example, global distance measurements may mistake non-disease hub proteins that have dense interactions with known disease proteins for potential disease proteins. To find a new method to avoid the aforementioned problem, we analyzed the differences between disease proteins and other proteins by using essential proteins (proteins encoded by essential genes) as references. We find that disease proteins are not well connected with essential proteins in the protein interaction networks. Based on this new finding, we proposed a novel strategy for gene prioritization based on protein interaction networks. We allocated positive flow to disease genes and negative flow to essential genes, and adopted network propagation for gene prioritization. Experimental results on 110 diseases verified the effectiveness and potential of the proposed method.

Laplacian normalization and random walk on heterogeneous networks for disease-gene prioritization

Random walk on heterogeneous networks is a recently emerging approach to effective disease gene prioritization. Laplacian normalization is a technique capable of normalizing the weight of edges in a network. We use this technique to normalize the gene matrix and the phenotype matrix before the construction of the heterogeneous network, and also use this idea to define the transition matrices of the heterogeneous network. Our method has remarkably better performance than the existing methods for recovering known gene-phenotype relationships. The Shannon information entropy of the distribution of the transition probabilities in our networks is found to be smaller than the networks constructed by the existing methods, implying that a higher number of top-ranked genes can be verified as disease genes. In fact, the most probable gene-phenotype relationships ranked within top 3 or top 5 in our gene lists can be confirmed by the OMIM database for many cases. Our algorithms have shown remarkably superior performance over the state-of-the-art algorithms for recovering gene-phenotype relationships. All Matlab codes can be available upon email request.

Network output controllability-based method for drug target identification

Biomolecules do not perform their functions alone, but interactively with one another to form so called biomolecular networks. It is well known that a complex disease stems from the malfunctions of corresponding biomolecular networks. Therefore, one of important tasks is to identify drug targets from biomolecular networks. In this study, the drug target identification is formulated as a problem of finding steering nodes in biomolecular networks while the concept of network output controllability is applied to the problem of drug target identification. By applying control signals to these steering nodes, the biomolecular networks are expected to be transited from one state to another. A graph-theoretic algorithm has been proposed to find a minimum set of steering nodes in biomolecular networks which can be a potential set of drug targets. Application results of the method to real biomolecular networks show that identified potential drug targets are in agreement with existing research results. This indicates that the method can generate testable predictions and provide insights into experimental design of drug discovery.

Performance of case-control rare copy number variation annotation in classification of autism

BACKGROUND

A substantial proportion of Autism Spectrum Disorder (ASD) risk resides in de novo germline and rare inherited genetic variation. In particular, rare copy number variation (CNV) contributes to ASD risk in up to 10% of ASD subjects. Despite the striking degree of genetic heterogeneity, case-control studies have detected specific burden of rare disruptive CNV for neuronal and neurodevelopmental pathways. Here, we used machine learning methods to classify ASD subjects and controls, based on rare CNV data and comprehensive gene annotations. We investigated performance of different methods and estimated the percentage of ASD subjects that could be reliably classified based on presumed etiologic CNV they carry.

RESULTS

We analyzed 1,892 Caucasian ASD subjects and 2,342 matched controls. Rare CNVs (frequency 1% or less) were detected using Illumina 1M and 1M-Duo BeadChips. Conditional Inference Forest (CF) typically performed as well as or better than other classification methods. We found a maximum AUC (area under the ROC curve) of 0.533 when considering all ASD subjects with rare genic CNVs, corresponding to 7.9% correctly classified ASD subjects and less than 3% incorrectly classified controls; performance was significantly higher when considering only subjects harboring de novo or pathogenic CNVs. We also found rare losses to be more predictive than gains and that curated neurally-relevant annotations (brain expression, synaptic components and neurodevelopmental phenotypes) outperform Gene Ontology and pathway-based annotations.

CONCLUSIONS

CF is an optimal classification approach for case-control rare CNV data and it can be used to prioritize subjects with variants potentially contributing to ASD risk not yet recognized. The neurally-relevant annotations used in this study could be successfully applied to rare CNV case-control data-sets for other neuropsychiatric disorders.

A disease module in the interactome explains disease heterogeneity, drug response and captures novel pathways and genes in asthma

Recent advances in genetics have spurred rapid progress towards the systematic identification of genes involved in complex diseases. Still, the detailed understanding of the molecular and physiological mechanisms through which these genes affect disease phenotypes remains a major challenge. Here, we identify the asthma disease module, i.e. the local neighborhood of the interactome whose perturbation is associated with asthma, and validate it for functional and pathophysiological relevance, using both computational and experimental approaches. We find that the asthma disease module is enriched with modest GWAS P-values against the background of random variation, and with differentially expressed genes from normal and asthmatic fibroblast cells treated with an asthma-specific drug. The asthma module also contains immune response mechanisms that are shared with other immune-related disease modules. Further, using diverse omics (genomics, gene-expression, drug response) data, we identify the GAB1 signaling pathway as an important novel modulator in asthma. The wiring diagram of the uncovered asthma module suggests a relatively close link between GAB1 and glucocorticoids (GCs), which we experimentally validate, observing an increase in the level of GAB1 after GC treatment in BEAS-2B bronchial epithelial cells. The siRNA knockdown of GAB1 in the BEAS-2B cell line resulted in a decrease in the NFkB level, suggesting a novel regulatory path of the pro-inflammatory factor NFkB by GAB1 in asthma.

Pathway mapping and development of disease-specific biomarkers: protein-based network biomarkers

It is known that a disease is rarely a consequence of an abnormality of a single gene, but reflects the interactions of various processes in a complex network. Annotated molecular networks offer new opportunities to understand diseases within a systems biology framework and provide an excellent substrate for network-based identification of biomarkers. The network biomarkers and dynamic network biomarkers (DNBs) represent new types of biomarkers with protein-protein or gene-gene interactions that can be monitored and evaluated at different stages and time-points during development of disease. Clinical bioinformatics as a new way to combine clinical measurements and signs with human tissue-generated bioinformatics is crucial to translate biomarkers into clinical application, validate the disease specificity, and understand the role of biomarkers in clinical settings. In this article, the recent advances and developments on network biomarkers and DNBs are comprehensively reviewed. How network biomarkers help a better understanding of molecular mechanism of diseases, the advantages and constraints of network biomarkers for clinical application, clinical bioinformatics as a bridge to the development of diseases-specific, stage-specific, severity-specific and therapy predictive biomarkers, and the potentials of network biomarkers are also discussed.

Discovery and characterization of functional modules and pathogenic genes associated with the risk of coronary artery disease

An integrated network biology approach for identifying disease risk functional modules and risk pathogenic genes for associated with CAD risk.

Systems Medicine: from molecular features and models to the clinic in COPD

Background and hypothesis

Chronic Obstructive Pulmonary Disease (COPD) patients are characterized by heterogeneous clinical manifestations and patterns of disease progression. Two major factors that can be used to identify COPD subtypes are muscle dysfunction/wasting and co-morbidity patterns. We hypothesized that COPD heterogeneity is in part the result of complex interactions between several genes and pathways. We explored the possibility of using a Systems Medicine approach to identify such pathways, as well as to generate predictive computational models that may be used in clinic practice.

Objective and method

Our overarching goal is to generate clinically applicable predictive models that characterize COPD heterogeneity through a Systems Medicine approach. To this end we have developed a general framework, consisting of three steps/objectives: (1) feature identification, (2) model generation and statistical validation, and (3) application and validation of the predictive models in the clinical scenario. We used muscle dysfunction and co-morbidity as test cases for this framework.

Results

In the study of muscle wasting we identified relevant features (genes) by a network analysis and generated predictive models that integrate mechanistic and probabilistic models. This allowed us to characterize muscle wasting as a general de-regulation of pathway interactions. In the co-morbidity analysis we identified relevant features (genes/pathways) by the integration of gene-disease and disease-disease associations. We further present a detailed characterization of co-morbidities in COPD patients that was implemented into a predictive model. In both use cases we were able to achieve predictive modeling but we also identified several key challenges, the most pressing being the validation and implementation into actual clinical practice.

Conclusions

The results confirm the potential of the Systems Medicine approach to study complex diseases and generate clinically relevant predictive models. Our study also highlights important obstacles and bottlenecks for such approaches (e.g. data availability and normalization of frameworks among others) and suggests specific proposals to overcome them.

Systematic prioritization and integrative analysis of copy number variations in schizophrenia reveal key schizophrenia susceptibility genes

Schizophrenia is a common mental disorder with high heritability and strong genetic heterogeneity. Common disease-common variants hypothesis predicts that schizophrenia is attributable in part to common genetic variants. However, recent studies have clearly demonstrated that copy number variations (CNVs) also play pivotal roles in schizophrenia susceptibility and explain a proportion of missing heritability. Though numerous CNVs have been identified, many of the regions affected by CNVs show poor overlapping among different studies, and it is not known whether the genes disrupted by CNVs contribute to the risk of schizophrenia. By using cumulative scoring, we systematically prioritized the genes affected by CNVs in schizophrenia. We identified 8 top genes that are frequently disrupted by CNVs, including NRXN1, CHRNA7, BCL9, CYFIP1, GJA8, NDE1, SNAP29, and GJA5. Integration of genes affected by CNVs with known schizophrenia susceptibility genes (from previous genetic linkage and association studies) reveals that many genes disrupted by CNVs are also associated with schizophrenia. Further protein-protein interaction (PPI) analysis indicates that protein products of genes affected by CNVs frequently interact with known schizophrenia-associated proteins. Finally, systematic integration of CNVs prioritization data with genetic association and PPI data identifies key schizophrenia candidate genes. Our results provide a global overview of genes impacted by CNVs in schizophrenia and reveal a densely interconnected molecular network of de novo CNVs in schizophrenia. Though the prioritized top genes represent promising schizophrenia risk genes, further work with different prioritization methods and independent samples is needed to confirm these findings. Nevertheless, the identified key candidate genes may have important roles in the pathogenesis of schizophrenia, and further functional characterization of these genes may provide pivotal targets for future therapeutics and diagnostics.

Visualization of genetic disease-phenotype similarities by multiple maps t-SNE with Laplacian regularization

BACKGROUND

From a phenotypic standpoint, certain types of diseases may prove to be difficult to accurately diagnose, due to specific combinations of confounding symptoms. Referred to as phenotypic overlap, these sets of disease-related symptoms suggest shared pathophysiological mechanisms. Few attempts have been made to visualize the phenotypic relationships between different human diseases from a machine learning perspective. The proposed research, it is anticipated, will visually assist researchers in quickly disambiguating symptoms which can confound the timely and accurate diagnosis of a disease.

METHODS

Our method is primarily based on multiple maps t-SNE (mm-tSNE), which is a probabilistic method for visualizing data points in multiple low dimensional spaces. We improved mm-tSNE by adding a Laplacian regularization term and subsequently provide an algorithm for optimizing the new objective function. The advantage of Laplacian regularization is that it adopts clustering structures of variables and provides more sparsity to the estimated parameters.

RESULTS

In order to further assess our modified mm-tSNE algorithm from a comparative standpoint, we reexamined two social network datasets used by the previous authors. Subsequently, we apply our method on phenotype dataset. In all these cases, our proposed method demonstrated better performance than the original version of mm-tSNE, as measured by the neighbourhood preservation ratio.

CONCLUSIONS

Phenotype grouping reflects the nature of human disease genetics. Thus, phenotype visualization may be complementary to investigate candidate genes for diseases as well as functional relations between genes and proteins. These relationships can be modelled by the modified mm-tSNE method. The modified mm-tSNE can be applied directly in other domain including social and biological datasets.

Prediction of disease-related genes based on weighted tissue-specific networks by using DNA methylation

BACKGROUND

Predicting disease-related genes is one of the most important tasks in bioinformatics and systems biology. With the advances in high-throughput techniques, a large number of protein-protein interactions are available, which make it possible to identify disease-related genes at the network level. However, network-based identification of disease-related genes is still a challenge as the considerable false-positives are still existed in the current available protein interaction networks (PIN).

RESULTS

Considering the fact that the majority of genetic disorders tend to manifest only in a single or a few tissues, we constructed tissue-specific networks (TSN) by integrating PIN and tissue-specific data. We further weighed the constructed tissue-specific network (WTSN) by using DNA methylation as it plays an irreplaceable role in the development of complex diseases. A PageRank-based method was developed to identify disease-related genes from the constructed networks. To validate the effectiveness of the proposed method, we constructed PIN, weighted PIN (WPIN), TSN, WTSN for colon cancer and leukemia, respectively. The experimental results on colon cancer and leukemia show that the combination of tissue-specific data and DNA methylation can help to identify disease-related genes more accurately. Moreover, the PageRank-based method was effective to predict disease-related genes on the case studies of colon cancer and leukemia.

CONCLUSIONS

Tissue-specific data and DNA methylation are two important factors to the study of human diseases. The same method implemented on the WTSN can achieve better results compared to those being implemented on original PIN, WPIN, or TSN. The PageRank-based method outperforms degree centrality-based method for identifying disease-related genes from WTSN.

Identifying disease genes by integrating multiple data sources

BACKGROUND

Now multiple types of data are available for identifying disease genes. Those data include gene-disease associations, disease phenotype similarities, protein-protein interactions, pathways, gene expression profiles, etc.. It is believed that integrating different kinds of biological data is an effective method to identify disease genes.

RESULTS

In this paper, we propose a multiple data integration method based on the theory of Markov random field (MRF) and the method of Bayesian analysis for identifying human disease genes. The proposed method is not only flexible in easily incorporating different kinds of data, but also reliable in predicting candidate disease genes.

CONCLUSIONS

Numerical experiments are carried out by integrating known gene-disease associations, protein complexes, protein-protein interactions, pathways and gene expression profiles. Predictions are evaluated by the leave-one-out method. The proposed method achieves an AUC score of 0.743 when integrating all those biological data in our experiments.

Network-based identification of biomarkers coexpressed with multiple pathways

Unraveling complex molecular interactions and networks and incorporating clinical information in modeling will present a paradigm shift in molecular medicine. Embedding biological relevance via modeling molecular networks and pathways has become increasingly important for biomarker identification in cancer susceptibility and metastasis studies. Here, we give a comprehensive overview of computational methods used for biomarker identification, and provide a performance comparison of several network models used in studies of cancer susceptibility, disease progression, and prognostication. Specifically, we evaluated implication networks, Boolean networks, Bayesian networks, and Pearson’s correlation networks in constructing gene coexpression networks for identifying lung cancer diagnostic and prognostic biomarkers. The results show that implication networks, implemented in Genet package, identified sets of biomarkers that generated an accurate prediction of lung cancer risk and metastases; meanwhile, implication networks revealed more biologically relevant molecular interactions than Boolean networks, Bayesian networks, and Pearson’s correlation networks when evaluated with MSigDB database.

Integration strategy is a key step in network-based analysis and dramatically affects network topological properties and inferring outcomes

An increasing number of experiments have been designed to detect intracellular and intercellular molecular interactions. Based on these molecular interactions (especially protein interactions), molecular networks have been built for using in several typical applications, such as the discovery of new disease genes and the identification of drug targets and molecular complexes. Because the data are incomplete and a considerable number of false-positive interactions exist, protein interactions from different sources are commonly integrated in network analyses to build a stable molecular network. Although various types of integration strategies are being applied in current studies, the topological properties of the networks from these different integration strategies, especially typical applications based on these network integration strategies, have not been rigorously evaluated. In this paper, systematic analyses were performed to evaluate 11 frequently used methods using two types of integration strategies: empirical and machine learning methods. The topological properties of the networks of these different integration strategies were found to significantly differ. Moreover, these networks were found to dramatically affect the outcomes of typical applications, such as disease gene predictions, drug target detections, and molecular complex identifications. The analysis presented in this paper could provide an important basis for future network-based biological researches.

Walking the interactome for candidate prioritization in exome sequencing studies of Mendelian diseases

Motivation: Whole-exome sequencing (WES) has opened up previously unheard of possibilities for identifying novel disease genes in Mendelian disorders, only about half of which have been elucidated to date. However, interpretation of WES data remains challenging.

Results: Here, we analyze protein–protein association (PPA) networks to identify candidate genes in the vicinity of genes previously implicated in a disease. The analysis, using a random-walk with restart (RWR) method, is adapted to the setting of WES by developing a composite variant-gene relevance score based on the rarity, location and predicted pathogenicity of variants and the RWR evaluation of genes harboring the variants. Benchmarking using known disease variants from 88 disease-gene families reveals that the correct gene is ranked among the top 10 candidates in ≥50% of cases, a figure which we confirmed using a prospective study of disease genes identified in 2012 and PPA data produced before that date. We implement our method in a freely available Web server, ExomeWalker, that displays a ranked list of candidates together with information on PPAs, frequency and predicted pathogenicity of the variants to allow quick and effective searches for candidates that are likely to reward closer investigation.

Availability and implementation: http://compbio.charite.de/ExomeWalker

Contact: peter.robinson@charite.de

Exploring mechanisms of human disease through structurally resolved protein interactome networks

The study of the molecular basis of human disease has gained increasing attention over the past decade. With significant improvements in sequencing efficiency and throughput, a wealth of genotypic data has become available. However the translation of this information into concrete advances in diagnostic and clinical setups has proved far more challenging. Two major reasons for this are the lack of functional annotation for genomic variants and the complex nature of genotype-to-phenotype relationships. One fundamental approach to bypass these issues is to examine the effects of genetic variation at the level of proteins as they are directly involved in carrying out biological functions. Within the cell, proteins function by interacting with other proteins as a part of an underlying interactome network. This network can be determined using interactome mapping - a combination of high-throughput experimental toolkits and curation from small-scale studies. Integrating structural information from co-crystals with the network allows generation of a structurally resolved network. Within the context of this network, the structural principles of disease mutations can be examined and used to generate reliable mechanistic hypotheses regarding disease pathogenesis.

Human symptoms-disease network

In the post-genomic era, the elucidation of the relationship between the molecular origins of diseases and their resulting phenotypes is a crucial task for medical research. Here, we use a large-scale biomedical literature database to construct a symptom-based human disease network and investigate the connection between clinical manifestations of diseases and their underlying molecular interactions. We find that the symptom-based similarity of two diseases correlates strongly with the number of shared genetic associations and the extent to which their associated proteins interact. Moreover, the diversity of the clinical manifestations of a disease can be related to the connectivity patterns of the underlying protein interaction network. The comprehensive, high-quality map of disease-symptom relations can further be used as a resource helping to address important questions in the field of systems medicine, for example, the identification of unexpected associations between diseases, disease etiology research or drug design.

A systems biology investigation of neurodegenerative dementia reveals a pivotal role of autophagy

Background

Neurodegenerative dementia comprises chronic and progressive illnesses with major clinical features represented by progressive and permanent loss of cognitive and mental performance, including impairment of memory and brain functions. Many different forms of neurodegenerative dementia exist, but they are all characterized by death of specific subpopulation of neurons and accumulation of proteins in the brain. We incorporated data from OMIM and primary molecular targets of drugs in the different phases of the drug discovery process to try to reveal possible hidden mechanism in neurodegenerative dementia. In the present study, a systems biology approach was used to investigate the molecular connections among seemingly distinct complex diseases with the shared clinical symptoms of dementia that could suggest related disease mechanisms.

Results

Network analysis was applied to characterize an interaction network of disease proteins and drug targets, revealing a major role of metabolism and, predominantly, of autophagy process in dementia and, particularly, in tauopathies. Different phases of the autophagy molecular pathway appear to be implicated in the individual disease pathophysiology and specific drug targets associated to autophagy modulation could be considered for pharmacological intervention. In particular, in view of their centrality and of the direct association to autophagy proteins in the network, PP2A subunits could be suggested as a suitable molecular target for the development of novel drugs.

Conclusion

The present systems biology investigation identifies the autophagy pathway as a central dis-regulated process in neurodegenerative dementia with a prevalent involvement in diseases characterized by tau inclusion and indicates the disease-specific molecules in the pathway that could be considered for therapy.

Protein-protein interactions and genetic diseases: The interactome

Protein-protein interactions mediate essentially all biological processes. Despite the quality of these data being widely questioned a decade ago, the reproducibility of large-scale protein interaction data is now much improved and there is little question that the latest screens are of high quality. Moreover, common data standards and coordinated curation practices between the databases that collect the interactions have made these valuable data available to a wide group of researchers. Here, I will review how protein-protein interactions are measured, collected and quality controlled. I discuss how the architecture of molecular protein networks has informed disease biology, and how these data are now being computationally integrated with the newest genomic technologies, in particular genome-wide association studies and exome-sequencing projects, to improve our understanding of molecular processes perturbed by genetics in human diseases. This article is part of a Special Issue entitled: From Genome to Function.

Network based integrated analysis of phenotype-genotype data for prioritization of candidate symptom genes

BACKGROUND

Symptoms and signs (symptoms in brief) are the essential clinical manifestations for individualized diagnosis and treatment in traditional Chinese medicine (TCM). To gain insights into the molecular mechanism of symptoms, we develop a computational approach to identify the candidate genes of symptoms.

METHODS

This paper presents a network-based approach for the integrated analysis of multiple phenotype-genotype data sources and the prediction of the prioritizing genes for the associated symptoms. The method first calculates the similarities between symptoms and diseases based on the symptom-disease relationships retrieved from the PubMed bibliographic database. Then the disease-gene associations and protein-protein interactions are utilized to construct a phenotype-genotype network. The PRINCE algorithm is finally used to rank the potential genes for the associated symptoms.

RESULTS

The proposed method gets reliable gene rank list with AUC (area under curve) 0.616 in classification. Some novel genes like CALCA, ESR1, and MTHFR were predicted to be associated with headache symptoms, which are not recorded in the benchmark data set, but have been reported in recent published literatures.

CONCLUSIONS

Our study demonstrated that by integrating phenotype-genotype relationships into a complex network framework it provides an effective approach to identify candidate genes of symptoms.

Computational approaches for human disease gene prediction and ranking

While candidate gene association studies continue to be the most practical and frequently employed approach in disease gene investigation for complex disorders, selecting suitable genes to test is a challenge. There are several computational approaches available for selecting and prioritizing disease candidate genes. A majority of these tools are based on guilt-by-association principle where novel disease candidate genes are identified and prioritized based on either functional or topological similarity to known disease genes. In this chapter we review the prioritization criteria and the algorithms along with some use cases that demonstrate how these tools can be used for identifying and ranking human disease candidate genes.

Analysis of the robustness of network-based disease-gene prioritization methods reveals redundancy in the human interactome and functional diversity of disease-genes

Complex biological systems usually pose a trade-off between robustness and fragility where a small number of perturbations can substantially disrupt the system. Although biological systems are robust against changes in many external and internal conditions, even a single mutation can perturb the system substantially, giving rise to a pathophenotype. Recent advances in identifying and analyzing the sequential variations beneath human disorders help to comprehend a systemic view of the mechanisms underlying various disease phenotypes. Network-based disease-gene prioritization methods rank the relevance of genes in a disease under the hypothesis that genes whose proteins interact with each other tend to exhibit similar phenotypes. In this study, we have tested the robustness of several network-based disease-gene prioritization methods with respect to the perturbations of the system using various disease phenotypes from the Online Mendelian Inheritance in Man database. These perturbations have been introduced either in the protein-protein interaction network or in the set of known disease-gene associations. As the network-based disease-gene prioritization methods are based on the connectivity between known disease-gene associations, we have further used these methods to categorize the pathophenotypes with respect to the recoverability of hidden disease-genes. Our results have suggested that, in general, disease-genes are connected through multiple paths in the human interactome. Moreover, even when these paths are disturbed, network-based prioritization can reveal hidden disease-gene associations in some pathophenotypes such as breast cancer, cardiomyopathy, diabetes, leukemia, parkinson disease and obesity to a greater extend compared to the rest of the pathophenotypes tested in this study. Gene Ontology (GO) analysis highlighted the role of functional diversity for such diseases.

Prioritization of candidate disease genes by enlarging the seed set and fusing information of the network topology and gene expression

The identification of disease genes is very important not only to provide greater understanding of gene function and cellular mechanisms which drive human disease, but also to enhance human disease diagnosis and treatment. Recently, high-throughput techniques have been applied to detect dozens or even hundreds of candidate genes. However, experimental approaches to validate the many candidates are usually time-consuming, tedious and expensive, and sometimes lack reproducibility. Therefore, numerous theoretical and computational methods (e.g. network-based approaches) have been developed to prioritize candidate disease genes. Many network-based approaches implicitly utilize the observation that genes causing the same or similar diseases tend to correlate with each other in gene-protein relationship networks. Of these network approaches, the random walk with restart algorithm (RWR) is considered to be a state-of-the-art approach. To further improve the performance of RWR, we propose a novel method named ESFSC to identify disease-related genes, by enlarging the seed set according to the centrality of disease genes in a network and fusing information of the protein-protein interaction (PPI) network topological similarity and the gene expression correlation. The ESFSC algorithm restarts at all of the nodes in the seed set consisting of the known disease genes and their k-nearest neighbor nodes, then walks in the global network separately guided by the similarity transition matrix constructed with PPI network topological similarity properties and the correlational transition matrix constructed with the gene expression profiles. As a result, all the genes in the network are ranked by weighted fusing the above results of the RWR guided by two types of transition matrices. Comprehensive simulation results of the 10 diseases with 97 known disease genes collected from the Online Mendelian Inheritance in Man (OMIM) database show that ESFSC outperforms existing methods for prioritizing candidate disease genes. The top prediction results of Alzheimer's disease are consistent with previous literature reports.

Approaches for recognizing disease genes based on network

Diseases are closely related to genes, thus indicating that genetic abnormalities may lead to certain diseases. The recognition of disease genes has long been a goal in biology, which may contribute to the improvement of health care and understanding gene functions, pathways, and interactions. However, few large-scale gene-gene association datasets, disease-disease association datasets, and gene-disease association datasets are available. A number of machine learning methods have been used to recognize disease genes based on networks. This paper states the relationship between disease and gene, summarizes the approaches used to recognize disease genes based on network, analyzes the core problems and challenges of the methods, and outlooks future research direction.

Identification of human disease genes from interactome network using graphlet interaction

Identifying genes related to human diseases, such as cancer and cardiovascular disease, etc., is an important task in biomedical research because of its applications in disease diagnosis and treatment. Interactome networks, especially protein-protein interaction networks, had been used to disease genes identification based on the hypothesis that strong candidate genes tend to closely relate to each other in some kinds of measure on the network. We proposed a new measure to analyze the relationship between network nodes which was called graphlet interaction. The graphlet interaction contained 28 different isomers. The results showed that the numbers of the graphlet interaction isomers between disease genes in interactome networks were significantly larger than random picked genes, while graphlet signatures were not. Then, we designed a new type of score, based on the network properties, to identify disease genes using graphlet interaction. The genes with higher scores were more likely to be disease genes, and all candidate genes were ranked according to their scores. Then the approach was evaluated by leave-one-out cross-validation. The precision of the current approach achieved 90% at about 10% recall, which was apparently higher than the previous three predominant algorithms, random walk, Endeavour and neighborhood based method. Finally, the approach was applied to predict new disease genes related to 4 common diseases, most of which were identified by other independent experimental researches. In conclusion, we demonstrate that the graphlet interaction is an effective tool to analyze the network properties of disease genes, and the scores calculated by graphlet interaction is more precise in identifying disease genes.

Network analysis of neurodegenerative disease highlights a role of Toll-like receptor signaling

Despite significant advances in the study of the molecular mechanisms altered in the development and progression of neurodegenerative diseases (NDs), the etiology is still enigmatic and the distinctions between diseases are not always entirely clear. We present an efficient computational method based on protein-protein interaction network (PPI) to model the functional network of NDs. The aim of this work is fourfold: (i) reconstruction of a PPI network relating to the NDs, (ii) construction of an association network between diseases based on proximity in the disease PPI network, (iii) quantification of disease associations, and (iv) inference of potential molecular mechanism involved in the diseases. The functional links of diseases not only showed overlap with the traditional classification in clinical settings, but also offered new insight into connections between diseases with limited clinical overlap. To gain an expanded view of the molecular mechanisms involved in NDs, both direct and indirect connector proteins were investigated. The method uncovered molecular relationships that are in common apparently distinct diseases and provided important insight into the molecular networks implicated in disease pathogenesis. In particular, the current analysis highlighted the Toll-like receptor signaling pathway as a potential candidate pathway to be targeted by therapy in neurodegeneration.

Network medicine analysis of chondrocyte proteins towards new treatments of osteoarthritis

Osteoarthritis (OA) is a progressive disorder with high incidence in the ageing human population that still has no treatment currently. This disorder induces the breakdown of articular cartilage, leading to the exposure and damage of bone surfaces. For a global understanding of OA development, the systematic integration of known OA-related proteins with protein-protein interaction (PPI) networks is required. In this work, the OA-related interactome was reconstructed using multiple data sources to have the most up-to-date information on OA-related proteins and their interactions. We then combined emergent concepts in network medicine to detect new unclassified OA-related proteins. The mapping of known OA-related proteins with PPI networks showed that these proteins are locally connected to each other and agglomerated in a large component. To expand this module, we applied a diffusion-based algorithm that probabilistically induces more searches in the vicinity of the seed OA-related proteins. As a result, the 10 topmost ranked proteins were connected to the OA disease module, supporting the local hypothesis. We computed structural modules and selected those that had the highest enrichment of OA-related proteins. The identified molecules show a link between structural topology and disease dysfunctionality. Interestingly, the protein Q6EEV6 was highlighted for OA association by both methods, reinforcing the potential involvement of this protein. These results suggest that similar disease-connected modules may exist in different human disorders, which could lead to systematic identification of genes or proteins that have a joint role in specific disease phenotypes.

Protein-protein interaction networks (PPI) and complex diseases

The physical interaction of proteins which lead to compiling them into large densely connected networks is a noticeable subject to investigation. Protein interaction networks are useful because of making basic scientific abstraction and improving biological and biomedical applications. Based on principle roles of proteins in biological function, their interactions determine molecular and cellular mechanisms, which control healthy and diseased states in organisms. Therefore, such networks facilitate the understanding of pathogenic (and physiologic) mechanisms that trigger the onset and progression of diseases. Consequently, this knowledge can be translated into effective diagnostic and therapeutic strategies. Furthermore, the results of several studies have proved that the structure and dynamics of protein networks are disturbed in complex diseases such as cancer and autoimmune disorders. Based on such relationship, a novel paradigm is suggested in order to confirm that the protein interaction networks can be the target of therapy for treatment of complex multi-genic diseases rather than individual molecules with disrespect the network.

Candidate gene identification for systemic lupus erythematosus using network centrality measures and gene ontology

Systemic lupus erythematosus (SLE) commonly accredited as “the great imitator” is a highly complex disease involving multiple gene susceptibility with non-specific symptoms. Many experimental and computational approaches have been used to investigate the disease related candidate genes. But the limited knowledge of gene function and disease correlation and also lack of complete functional details about the majority of genes in susceptible locus, encumbrances the identification of SLE related candidate genes. In this paper, we have studied the human immunome network (undirected) using various graph theoretical centrality measures integrated with the gene ontology terms to predict the new candidate genes. As a result, we have identified 8 candidate genes, which may act as potential targets for SLE disease. We have also carried out the same analysis by replacing the human immunome network with human immunome signaling network (directed) and as an outcome we have obtained 5 candidate genes as potential targets for SLE disease. From the comparison study, we have found these two approaches are complementary in nature.

Network analysis of GWAS data

Genome-wide association studies (GWAS) identify genetic variants that distinguish a control population from a population with a specific trait. Two challenges in GWAS are: (1) identification of the causal variant within a longer haplotype that is associated with the trait; (2) identification of causal variants for polygenic traits that are caused by variants in multiple genes within a pathway. We review recent methods that use information in protein-protein and protein-DNA interaction networks to address these two challenges.

Controllability in cancer metabolic networks according to drug targets as driver nodes

Networks are employed to represent many nonlinear complex systems in the real world. The topological aspects and relationships between the structure and function of biological networks have been widely studied in the past few decades. However dynamic and control features of complex networks have not been widely researched, in comparison to topological network features. In this study, we explore the relationship between network controllability, topological parameters, and network medicine (metabolic drug targets). Considering the assumption that targets of approved anticancer metabolic drugs are driver nodes (which control cancer metabolic networks), we have applied topological analysis to genome-scale metabolic models of 15 normal and corresponding cancer cell types. The results show that besides primary network parameters, more complex network metrics such as motifs and clusters may also be appropriate for controlling the systems providing the controllability relationship between topological parameters and drug targets. Consequently, this study reveals the possibilities of following a set of driver nodes in network clusters instead of considering them individually according to their centralities. This outcome suggests considering distributed control systems instead of nodal control for cancer metabolic networks, leading to a new strategy in the field of network medicine.

An algorithm for network-based gene prioritization that encodes knowledge both in nodes and in links

BACKGROUND

Candidate gene prioritization aims to identify promising new genes associated with a disease or a biological process from a larger set of candidate genes. In recent years, network-based methods - which utilize a knowledge network derived from biological knowledge - have been utilized for gene prioritization. Biological knowledge can be encoded either through the network's links or nodes. Current network-based methods can only encode knowledge through links. This paper describes a new network-based method that can encode knowledge in links as well as in nodes.

RESULTS

We developed a new network inference algorithm called the Knowledge Network Gene Prioritization (KNGP) algorithm which can incorporate both link and node knowledge. The performance of the KNGP algorithm was evaluated on both synthetic networks and on networks incorporating biological knowledge. The results showed that the combination of link knowledge and node knowledge provided a significant benefit across 19 experimental diseases over using link knowledge alone or node knowledge alone.

CONCLUSIONS

The KNGP algorithm provides an advance over current network-based algorithms, because the algorithm can encode both link and node knowledge. We hope the algorithm will aid researchers with gene prioritization.