Enhancing the prioritization of disease-causing genes through tissue specific protein interaction networks

SPIDER: constructing cell-type-specific protein-protein interaction networks

Motivation

Protein-protein interactions (PPIs) play essential roles in the buildup of cellular machinery and provide the skeleton for cellular signaling. However, these biochemical roles are context dependent and interactions may change across cell type, time, and space. In contrast, PPI detection assays are run in a single condition that may not even be an endogenous condition of the organism, resulting in static networks that do not reflect full cellular complexity. Thus, there is a need for computational methods to predict cell-type-specific interactions.

Results

Here we present SPIDER (Supervised Protein Interaction DEtectoR), a graph attention-based model for predicting cell-type-specific PPI networks. In contrast to previous attempts at this problem, which were unsupervised in nature, our model's training is guided by experimentally measured cell-type-specific networks, enhancing its performance. We evaluate our method using experimental data of cell-type-specific networks from both humans and mice, and show that it outperforms current approaches by a large margin. We further demonstrate the ability of our method to generalize the predictions to datasets of tissues lacking prior PPI experimental data. We leverage the networks predicted by the model to facilitate the identification of tissue-specific disease genes.

Availability and implementation

Our code and data are available at https://github.com/Kuper994/SPIDER.

State of the Interactomes: an evaluation of molecular networks for generating biological insights

Advancements in genomic and proteomic technologies have powered the use of gene and protein networks ("interactomes") for understanding genotype-phenotype translation. However, the proliferation of interactomes complicates the selection of networks for specific applications. Here, we present a comprehensive evaluation of 46 current human interactomes, encompassing protein-protein interactions as well as gene regulatory, signaling, colocalization, and genetic interaction networks. Our analysis shows that large composite networks such as HumanNet, STRING, and FunCoup are most effective for identifying disease genes, while smaller networks such as DIP and SIGNOR demonstrate strong interaction prediction performance. These findings provide a benchmark for interactomes across diverse network biology applications and clarify factors that influence network performance. Furthermore, our evaluation pipeline paves the way for continued assessment of emerging and updated interaction networks in the future.

Neurogenetic underpinnings of nicotine use severity: Integrating the brain transcriptomes and GWAS variants via network approaches

Our study focused on human brain transcriptomes and the genetic risks of cigarettes per day (CPD) to investigate the neurogenetic mechanisms of individual variation in nicotine use severity. We constructed whole-brain and intramodular region-specific coexpression networks using BrainSpan's transcriptomes, and the genomewide association studies identified risk variants of CPD, confirmed the associations between CPD and each gene set in the region-specific subnetworks using an independent dataset, and conducted bioinformatic analyses. Eight brain-region-specific coexpression subnetworks were identified in association with CPD: amygdala, hippocampus, medial prefrontal cortex (MPFC), orbitofrontal cortex (OPFC), dorsolateral prefrontal cortex, striatum, mediodorsal nucleus of the thalamus (MDTHAL), and primary motor cortex (M1C). Each gene set in the eight subnetworks was associated with CPD. We also identified three hub proteins encoded by GRIN2A in the amygdala, PMCA2 in the hippocampus, MPFC, OPFC, striatum, and MDTHAL, and SV2B in M1C. Intriguingly, the pancreatic secretion pathway appeared in all the significant protein interaction subnetworks, suggesting pleiotropic effects between cigarette smoking and pancreatic diseases. The three hub proteins and genes are implicated in stress response, drug memory, calcium homeostasis, and inhibitory control. These findings provide novel evidence of the neurogenetic underpinnings of smoking severity.

Prioritization of genes for translation: a computational approach

INTRODUCTION

Gene identification for genetic diseases is critical for the development of new diagnostic approaches and personalized treatment options. Prioritization of gene translation is an important consideration in the molecular biology field, allowing researchers to focus on the most promising candidates for further investigation.

AREAS COVERED

In this paper, we discussed different approaches to prioritize genes for translation, including the use of computational tools and machine learning algorithms, as well as experimental techniques such as knockdown and overexpression studies. We also explored the potential biases and limitations of these approaches and proposed strategies to improve the accuracy and reliability of gene prioritization methods. Although numerous computational methods have been developed for this purpose, there is a need for computational methods that incorporate tissue-specific information to enable more accurate prioritization of candidate genes. Such methods should provide tissue-specific predictions, insights into underlying disease mechanisms, and more accurate prioritization of genes.

EXPERT OPINION

Using advanced computational tools and machine learning algorithms to prioritize genes, we can identify potential targets for therapeutic intervention of complex diseases. This represents an up-and-coming method for drug development and personalized medicine.

In vivo mapping of protein-protein interactions of schizophrenia risk factors generates an interconnected disease network

Genetic analyses of Schizophrenia (SCZ) patients have identified thousands of risk factors. In silico protein-protein interaction (PPI) network analysis has provided strong evidence that disrupted PPI networks underlie SCZ pathogenesis. In this study, we performed in vivo PPI analysis of several SCZ risk factors in the rodent brain. Using endogenous antibody immunoprecipitations coupled to mass spectrometry (MS) analysis, we constructed a SCZ network comprising 1612 unique PPI with a 5% FDR. Over 90% of the PPI were novel, reflecting the lack of previous PPI MS studies in brain tissue. Our SCZ PPI network was enriched with known SCZ risk factors, which supports the hypothesis that an accumulation of disturbances in selected PPI networks underlies SCZ. We used Stable Isotope Labeling in Mammals (SILAM) to quantitate phencyclidine (PCP) perturbations in the SCZ network and found that PCP weakened most PPI but also led to some enhanced or new PPI. These findings demonstrate that quantitating PPI in perturbed biological states can reveal alterations to network biology.

Predicting molecular mechanisms of hereditary diseases by using their tissue-selective manifestation

How do aberrations in widely expressed genes lead to tissue-selective hereditary diseases? Previous attempts to answer this question were limited to testing a few candidate mechanisms. To answer this question at a larger scale, we developed "Tissue Risk Assessment of Causality by Expression" (TRACE), a machine learning approach to predict genes that underlie tissue-selective diseases and selectivity-related features. TRACE utilized 4,744 biologically interpretable tissue-specific gene features that were inferred from heterogeneous omics datasets. Application of TRACE to 1,031 disease genes uncovered known and novel selectivity-related features, the most common of which was previously overlooked. Next, we created a catalog of tissue-associated risks for 18,927 protein-coding genes (https://netbio.bgu.ac.il/trace/). As proof-of-concept, we prioritized candidate disease genes identified in 48 rare-disease patients. TRACE ranked the verified disease gene among the patient's candidate genes significantly better than gene prioritization methods that rank by gene constraint or tissue expression. Thus, tissue selectivity combined with machine learning enhances genetic and clinical understanding of hereditary diseases.

Effective Subject Representation based on Multi-omics Disease Networks using Graph Embedding

The study of complex behavior of biological systems has become increasingly dependent on evolutionary network modeling. In particular, multi-omics networks capture interactions between biomolecules such as proteins and metabolites, providing a basis for predicting relationships between such biomolecules and various phenotypic traits of complex diseases. In this paper, we introduce an integrative framework that given a multi-omics network representing a cohort of subjects, learns expressive representations for network nodes, and combines the learned nodes representations with the biological profiles of individual subjects for enriched representation of the subjects. With extensive empirical evaluation using real-world multi-omics networks, we show that our proposed framework significantly outperforms existing and baseline methods in terms of subject representation accuracy, particularly when the multi-omics network representing the cohort is sparse and structured and therefore, more informative.

Gene co-expression analyses of health(span) across multiple species

Health(span)-related gene clusters/modules were recently identified based on knowledge about the cross-species genetic basis of health, to interpret transcriptomic datasets describing health-related interventions. However, the cross-species comparison of health-related observations reveals a lot of heterogeneity, not least due to widely varying health(span) definitions and study designs, posing a challenge for the exploration of conserved healthspan modules and, specifically, their transfer across species. To improve the identification and exploration of conserved/transferable healthspan modules, here we apply an established workflow based on gene co-expression network analyses employing GEO/ArrayExpress data for human and animal models, and perform a comprehensive meta-study of the resulting modules related to health(span), yielding a small set of literature backed health(span) candidate genes. For each experiment, WGCNA (weighted gene correlation network analysis) was used to infer modules of genes which correlate in their expression with a 'health phenotype score' and to determine the most-connected (hub) genes (and their interactions) for each such module. After mapping these hub genes to their human orthologs, 12 health(span) genes were identified in at least two species (ACTN3, ANK1, MRPL18, MYL1, PAXIP1, PPP1CA, SCN3B, SDCBP, SKIV2L, TUBG1, TYROBP, WIPF1), for which enrichment analysis by g:profiler found an association with actin filament-based movement and associated organelles, as well as muscular structures. We conclude that a meta-study of hub genes from co-expression network analyses for the complex phenotype health(span), across multiple species, can yield molecular-mechanistic insights and can direct experimentalists to further investigate the contribution of individual genes and their interactions to health(span).

Distinct translatome changes in specific neural populations precede electroencephalographic changes in prion-infected mice

Selective vulnerability is an enigmatic feature of neurodegenerative diseases (NDs), whereby a widely expressed protein causes lesions in specific cell types and brain regions. Using the RiboTag method in mice, translational responses of five neural subtypes to acquired prion disease (PrD) were measured. Pre-onset and disease onset timepoints were chosen based on longitudinal electroencephalography (EEG) that revealed a gradual increase in theta power between 10- and 18-weeks after prion injection, resembling a clinical feature of human PrD. At disease onset, marked by significantly increased theta power and histopathological lesions, mice had pronounced translatome changes in all five cell types despite appearing normal. Remarkably, at a pre-onset stage, prior to EEG and neuropathological changes, we found that 1) translatomes of astrocytes indicated reduced synthesis of ribosomal and mitochondrial components, 2) glutamatergic neurons showed increased expression of cytoskeletal genes, and 3) GABAergic neurons revealed reduced expression of circadian rhythm genes. These data demonstrate that early translatome responses to neurodegeneration emerge prior to conventional markers of disease and are cell type-specific. Therapeutic strategies may need to target multiple pathways in specific populations of cells, early in disease.

Prions are infectious agents composed of a misfolded protein. When isolated from a mammalian brain and transferred to the same host species, prions will cause the same neurodegenerative disease affecting the same brain regions and cell types. This concept of selective vulnerability is also a feature of more common types of neurodegenerative diseases, such as Alzheimer’s, Parkinson’s, and Huntington’s. To better understand the mechanisms behind selective vulnerability, we studied disease responses of five cell types with different vulnerabilities in prion-infected mice at two different disease stages. Responses were measured as changes to mRNAs undergoing translation, referred to as the translatome. Before prion-infected mice demonstrated typical disease signs, electroencephalography (a method used clinically to characterize neurodegeneration in humans) revealed brain changes resembling those in human prion diseases, and surprisingly, the translatomes of all cells were drastically changed. Furthermore, before electroencephalography changes emerged, three cell types made unique responses while the most vulnerable cell type did not. These results suggests that mechanisms causing selective vulnerability will be difficult to dissect and that therapies will likely need to be provided before clinical signs emerge and individually engage multiple cell types and their distinct molecular pathways.

A high-resolution map of human RNA translation

Translated small open reading frames (smORFs) can have important regulatory roles and encode microproteins, yet their genome-wide identification has been challenging. We determined the ribosome locations across six primary human cell types and five tissues and detected 7,767 smORFs with translational profiles matching those of known proteins. The human genome was found to contain highly cell-type- and tissue-specific smORFs and a subset that encodes highly conserved amino acid sequences. Changes in the translational efficiency of upstream-encoded smORFs (uORFs) and the corresponding main ORFs predominantly occur in the same direction. Integration with 456 mass-spectrometry datasets confirms the presence of 603 small peptides at the protein level in humans and provides insights into the subcellular localization of these small proteins. This study provides a comprehensive atlas of high-confidence translated smORFs derived from primary human cells and tissues in order to provide a more complete understanding of the translated human genome.

GLIDER: Function Prediction from GLIDE-based Neigborhoods

MOTIVATION

Protein function prediction, based on the patterns of connection in a Protein-Protein Interaction (or Association) network, is perhaps the most studied of the classical, fundamental inference problems for biological networks. A highly successful set of recent approaches use random walk-based low dimensional embeddings, that tend to place functionally similar proteins into coherent spatial regions. However, these approaches lose valuable local graph structure from the network when considering only the embedding. We introduce GLIDER, a method that replaces a protein-protein interaction or association network with a new graph-based similarity network. GLIDER is based on a variant of our previous GLIDE method, which was designed to predict missing links in Protein-Protein Association networks, capturing implicit local and global (i.e. embedding-based) graph properties.

RESULTS

GLIDER outperforms competing methods on the task of predicting GO functional labels in cross-validation on a heterogeneous collection of four Human Protein-Protein Association networks derived from the 2016 DREAM Disease Module Identification Challenge, and also on three different protein-protein association networks built from the STRING database. We show that this is due to the strong functional enrichment that is present in the local GLIDER neighborhood in multiple different types of protein-protein association networks. Furthermore, we introduce the GLIDER graph neighborhood as a way for biologists to visualize the local neighborhood of a disease gene. As an application, we look at the local GLIDER neighborhoods of a set of known Parkinson's Disease GWAS genes, rediscover many genes which have known involvement in Parkinson's disease pathways, plus suggest some new genes to study.

AVAILABILITY

All code is publicly available and can be accessed here: https://github.com/kap-devkota/GLIDER.

SUPPLEMENTARY INFORMATION

is available at Bioinformatics online.

Transcription factor protein interactomes reveal genetic determinants in heart disease

Congenital heart disease (CHD) is present in 1% of live births, yet identification of causal mutations remains challenging. We hypothesized that genetic determinants for CHDs may lie in the protein interactomes of transcription factors whose mutations cause CHDs. Defining the interactomes of two transcription factors haplo-insufficient in CHD, GATA4 and TBX5, within human cardiac progenitors, and integrating the results with nearly 9,000 exomes from proband-parent trios revealed an enrichment of de novo missense variants associated with CHD within the interactomes. Scoring variants of interactome members based on residue, gene, and proband features identified likely CHD-causing genes, including the epigenetic reader GLYR1. GLYR1 and GATA4 widely co-occupied and co-activated cardiac developmental genes, and the identified GLYR1 missense variant disrupted interaction with GATA4, impairing in vitro and in vivo function in mice. This integrative proteomic and genetic approach provides a framework for prioritizing and interrogating genetic variants in heart disease.

Construction and contextualization approaches for protein-protein interaction networks

Protein-protein interaction network (PPIN) analysis is a widely used method to study the contextual role of proteins of interest, to predict novel disease genes, disease or functional modules, and to identify novel drug targets. PPIN-based analysis uses both generic and context-specific networks. Multiple contextualization methodologies have been described, such as shortest-path algorithms, neighborhood-based methods, and diffusion/propagation algorithms. This review discusses these methods, provides intuitive representations of PPIN contextualization, and also examines how the quality of such context-specific networks could be improved by considering additional sources of evidence. As a heuristic, we observe that tasks such as identifying disease genes, drug targets, and protein complexes should consider local neighborhoods, while uncovering disease mechanisms and discovering disease-pathways would gain from diffusion-based construction.

IID 2021: towards context-specific protein interaction analyses by increased coverage, enhanced annotation and enrichment analysis

Improved bioassays have significantly increased the rate of identifying new protein-protein interactions (PPIs), and the number of detected human PPIs has greatly exceeded early estimates of human interactome size. These new PPIs provide a more complete view of disease mechanisms but precise understanding of how PPIs affect phenotype remains a challenge. It requires knowledge of PPI context (e.g. tissues, subcellular localizations), and functional roles, especially within pathways and protein complexes. The previous IID release focused on PPI context, providing networks with comprehensive tissue, disease, cellular localization, and druggability annotations. The current update adds developmental stages to the available contexts, and provides a way of assigning context to PPIs that could not be previously annotated due to insufficient data or incompatibility with available context categories (e.g. interactions between membrane and cytoplasmic proteins). This update also annotates PPIs with conservation across species, directionality in pathways, membership in large complexes, interaction stability (i.e. stable or transient), and mutation effects. Enrichment analysis is now available for all annotations, and includes multiple options; for example, context annotations can be analyzed with respect to PPIs or network proteins. In addition to tabular view or download, IID provides online network visualization. This update is available at http://ophid.utoronto.ca/iid.

Decoding disease: from genomes to networks to phenotypes

Interpreting the effects of genetic variants is key to understanding individual susceptibility to disease and designing personalized therapeutic approaches. Modern experimental technologies are enabling the generation of massive compendia of human genome sequence data and associated molecular and phenotypic traits, together with genome-scale expression, epigenomics and other functional genomic data. Integrative computational models can leverage these data to understand variant impact, elucidate the effect of dysregulated genes on biological pathways in specific disease and tissue contexts, and interpret disease risk beyond what is feasible with experiments alone. In this Review, we discuss recent developments in machine learning algorithms for genome interpretation and for integrative molecular-level modelling of cells, tissues and organs relevant to disease. More specifically, we highlight existing methods and key challenges and opportunities in identifying specific disease-causing genetic variants and linking them to molecular pathways and, ultimately, to disease phenotypes.

An atlas of protein-protein interactions across mouse tissues

Cellular processes arise from the dynamic organization of proteins in networks of physical interactions. Mapping the interactome has therefore been a central objective of high-throughput biology. However, the dynamics of protein interactions across physiological contexts remain poorly understood. Here, we develop a quantitative proteomic approach combining protein correlation profiling with stable isotope labeling of mammals (PCP-SILAM) to map the interactomes of seven mouse tissues. The resulting maps provide a proteome-scale survey of interactome rewiring across mammalian tissues, revealing more than 125,000 unique interactions at a quality comparable to the highest-quality human screens. We identify systematic suppression of cross-talk between the evolutionarily ancient housekeeping interactome and younger, tissue-specific modules. Rewired proteins are tightly regulated by multiple cellular mechanisms and are implicated in disease. Our study opens up new avenues to uncover regulatory mechanisms that shape in vivo interactome responses to physiological and pathophysiological stimuli in mammalian systems.

Optimized Molecular Interaction Networks for the Study of Skeletal Muscle

BACKGROUND

Molecular interaction networks (MINs) aim to capture the complex relationships between interacting molecules within a biological system. MINs can be constructed from existing knowledge of molecular functional associations, such as protein-protein binding interactions (PPI) or gene co-expression, and these different sources may be combined into a single MIN. A given MIN may be more or less optimal in its representation of the important functional relationships of molecules in a tissue.

OBJECTIVE

The aim of this study was to establish whether a combined MIN derived from different types of functional association could better capture muscle-relevant biology compared to its constituent single-source MINs.

METHODS

MINs were constructed from functional association databases for both protein-binding and gene co-expression. The networks were then compared based on the capture of muscle-relevant genes and gene ontology (GO) terms, tested in two different ways using established biological network clustering algorithms. The top performing MINs were combined to test whether an optimal MIN for skeletal muscle could be constructed.

RESULTS

The STRING PPI network was the best performing single-source MIN among those tested. Combining STRING with interactions from either the MyoMiner or CoXPRESSdb gene co-expression sources resulted in a combined network with improved performance relative to its constituent networks.

CONCLUSION

MINs constructed from multiple types of functional association can better represent the functional relationships of molecules in a given tissue. Such networks may be used to improve the analysis and interpretation of functional genomics data in the study of skeletal muscle and neuromuscular diseases. Networks and clusters described by this study, including the combinations of STRING with MyoMiner or with CoXPRESSdb, are available for download from https://www.sys-myo.com/myominer/download.php.

Disease gene prediction with privileged information and heteroscedastic dropout

Motivation

Recently, machine learning models have achieved tremendous success in prioritizing candidate genes for genetic diseases. These models are able to accurately quantify the similarity among disease and genes based on the intuition that similar genes are more likely to be associated with similar diseases. However, the genetic features these methods rely on are often hard to collect due to high experimental cost and various other technical limitations. Existing solutions of this problem significantly increase the risk of overfitting and decrease the generalizability of the models.

Results

In this work, we propose a graph neural network (GNN) version of the Learning under Privileged Information paradigm to predict new disease gene associations. Unlike previous gene prioritization approaches, our model does not require the genetic features to be the same at training and test stages. If a genetic feature is hard to measure and therefore missing at the test stage, our model could still efficiently incorporate its information during the training process. To implement this, we develop a Heteroscedastic Gaussian Dropout algorithm, where the dropout probability of the GNN model is determined by another GNN model with a mirrored GNN architecture. To evaluate our method, we compared our method with four state-of-the-art methods on the Online Mendelian Inheritance in Man dataset to prioritize candidate disease genes. Extensive evaluations show that our model could improve the prediction accuracy when all the features are available compared to other methods. More importantly, our model could make very accurate predictions when >90% of the features are missing at the test stage.

Availability and implementation

Our method is realized with Python 3.7 and Pytorch 1.5.0 and method and data are freely available at: https://github.com/juanshu30/Disease-Gene-Prioritization-with-Privileged-Information-and-Heteroscedastic-Dropout.

Analysing omics data sets with weighted nodes networks (WNNets)

Current trends in biomedical research indicate data integration as a fundamental step towards precision medicine. In this context, network models allow representing and analysing complex biological processes. However, although effective in unveiling network properties, these models fail in considering the individual, biochemical variations occurring at molecular level. As a consequence, the analysis of these models partially loses its predictive power. To overcome these limitations, Weighted Nodes Networks (WNNets) were developed. WNNets allow to easily and effectively weigh nodes using experimental information from multiple conditions. In this study, the characteristics of WNNets were described and a proteomics data set was modelled and analysed. Results suggested that degree, an established centrality index, may offer a novel perspective about the functional role of nodes in WNNets. Indeed, degree allowed retrieving significant differences between experimental conditions, highlighting relevant proteins, and provided a novel interpretation for degree itself, opening new perspectives in experimental data modelling and analysis. Overall, WNNets may be used to model any high-throughput experimental data set requiring weighted nodes. Finally, improving the power of the analysis by using centralities such as betweenness may provide further biological insights and unveil novel, interesting characteristics of WNNets.

Predicting disease‐associated genes: Computational methods, databases, and evaluations

Complex diseases are associated with a set of genes (called disease genes), the identification of which can help scientists uncover the mechanisms of diseases and develop new drugs and treatment strategies. Due to the huge cost and time of experimental identification techniques, many computational algorithms have been proposed to predict disease genes. Although several review publications in recent years have discussed many computational methods, some of them focus on cancer driver genes while others focus on biomolecular networks, which only cover a specific aspect of existing methods. In this review, we summarize existing methods and classify them into three categories based on their rationales. Then, the algorithms, biological data, and evaluation methods used in the computational prediction are discussed. Finally, we highlight the limitations of existing methods and point out some future directions for improving these algorithms. This review could help investigators understand the principles of existing methods, and thus develop new methods to advance the computational prediction of disease genes.

This article is categorized under:

e-MutPath: computational modeling reveals the functional landscape of genetic mutations rewiring interactome networks

Understanding the functional impact of cancer somatic mutations represents a critical knowledge gap for implementing precision oncology. It has been increasingly appreciated that the interaction profile mediated by a genomic mutation provides a fundamental link between genotype and phenotype. However, specific effects on biological signaling networks for the majority of mutations are largely unknown by experimental approaches. To resolve this challenge, we developed e-MutPath (edgetic Mutation-mediated Pathway perturbations), a network-based computational method to identify candidate ‘edgetic’ mutations that perturb functional pathways. e-MutPath identifies informative paths that could be used to distinguish disease risk factors from neutral elements and to stratify disease subtypes with clinical relevance. The predicted targets are enriched in cancer vulnerability genes, known drug targets but depleted for proteins associated with side effects, demonstrating the power of network-based strategies to investigate the functional impact and perturbation profiles of genomic mutations. Together, e-MutPath represents a robust computational tool to systematically assign functions to genetic mutations, especially in the context of their specific pathway perturbation effect.

PINA 3.0: mining cancer interactome

Protein–protein interactions (PPIs) are crucial to mediate biological functions, and understanding PPIs in cancer type-specific context could help decipher the underlying molecular mechanisms of tumorigenesis and identify potential therapeutic options. Therefore, we update the Protein Interaction Network Analysis (PINA) platform to version 3.0, to integrate the unified human interactome with RNA-seq transcriptomes and mass spectrometry-based proteomes across tens of cancer types. A number of new analytical utilities were developed to help characterize the cancer context for a PPI network, which includes inferring proteins with expression specificity and identifying candidate prognosis biomarkers, putative cancer drivers, and therapeutic targets for a specific cancer type; as well as identifying pairs of co-expressing interacting proteins across cancer types. Furthermore, a brand-new web interface has been designed to integrate these new utilities within an interactive network visualization environment, which allows users to quickly and comprehensively investigate the roles of human interacting proteins in a cancer type-specific context. PINA is freely available at https://omics.bjcancer.org/pina/.

Network propagation of rare variants in Alzheimer's disease reveals tissue-specific hub genes and communities

State-of-the-art rare variant association testing methods aggregate the contribution of rare variants in biologically relevant genomic regions to boost statistical power. However, testing single genes separately does not consider the complex interaction landscape of genes, nor the downstream effects of non-synonymous variants on protein structure and function. Here we present the NETwork Propagation-based Assessment of Genetic Events (NETPAGE), an integrative approach aimed at investigating the biological pathways through which rare variation results in complex disease phenotypes. We applied NETPAGE to sporadic, late-onset Alzheimer's disease (AD), using whole-genome sequencing from the AD Neuroimaging Initiative (ADNI) cohort, as well as whole-exome sequencing from the AD Sequencing Project (ADSP). NETPAGE is based on network propagation, a framework that models information flow on a graph and simulates the percolation of genetic variation through tissue-specific gene interaction networks. The result of network propagation is a set of smoothed gene scores that can be tested for association with disease status through sparse regression. The application of NETPAGE to AD enabled the identification of a set of connected genes whose smoothed variation profile was robustly associated to case-control status, based on gene interactions in the hippocampus. Additionally, smoothed scores significantly correlated with risk of conversion to AD in Mild Cognitive Impairment (MCI) subjects. Lastly, we investigated tissue-specific transcriptional dysregulation of the core genes in two independent RNA-seq datasets, as well as significant enrichments in terms of gene sets with known connections to AD. We present a framework that enables enhanced genetic association testing for a wide range of traits, diseases, and sample sizes.

Differential Co-Expression Analyses Allow the Identification of Critical Signalling Pathways Altered during Tumour Transformation and Progression

Biological systems respond to perturbations through the rewiring of molecular interactions, organised in gene regulatory networks (GRNs). Among these, the increasingly high availability of transcriptomic data makes gene co-expression networks the most exploited ones. Differential co-expression networks are useful tools to identify changes in response to an external perturbation, such as mutations predisposing to cancer development, and leading to changes in the activity of gene expression regulators or signalling. They can help explain the robustness of cancer cells to perturbations and identify promising candidates for targeted therapy, moreover providing higher specificity with respect to standard co-expression methods. Here, we comprehensively review the literature about the methods developed to assess differential co-expression and their applications to cancer biology. Via the comparison of normal and diseased conditions and of different tumour stages, studies based on these methods led to the definition of pathways involved in gene network reorganisation upon oncogenes' mutations and tumour progression, often converging on immune system signalling. A relevant implementation still lagging behind is the integration of different data types, which would greatly improve network interpretability. Most importantly, performance and predictivity evaluation of the large variety of mathematical models proposed would urgently require experimental validations and systematic comparisons. We believe that future work on differential gene co-expression networks, complemented with additional omics data and experimentally tested, will considerably improve our insights into the biology of tumours.

Rank-Similarity Measures for Comparing Gene Prioritizations: A Case Study in Autism

We discuss the challenge of comparing three gene prioritization methods: network propagation, integer linear programming rank aggregation (RA), and statistical RA. These methods are based on different biological categories and estimate disease-gene association. Previously proposed comparison schemes are based on three measures of performance: receiver operating curve, area under the curve, and median rank ratio. Although they may capture important aspects of gene prioritization performance, they may fail to capture important differences in the rankings of individual genes. We suggest that comparison schemes could be improved by also considering recently proposed measures of similarity between gene rankings. We tested this suggestion on comparison schemes for prioritizations of genes associated with autism that were obtained using brain- and tissue-specific data. Our results show the effectiveness of our measures of similarity in clustering brain regions based on their relevance to autism.

Exploring Drug Treatment Patterns Based on the Action of Drug and Multilayer Network Model

Some drugs can be used to treat multiple diseases, suggesting potential patterns in drug treatment. Determination of drug treatment patterns can improve our understanding of the mechanisms of drug action, enabling drug repurposing. A drug can be associated with a multilayer tissue-specific protein-protein interaction (TSPPI) network for the diseases it is used to treat. Proteins usually interact with other proteins to achieve functions that cause diseases. Hence, studying drug treatment patterns is similar to studying common module structures in multilayer TSPPI networks. Therefore, we propose a network-based model to study the treatment patterns of drugs. The method was designated SDTP (studying drug treatment pattern) and was based on drug effects and a multilayer network model. To demonstrate the application of the SDTP method, we focused on analysis of trichostatin A (TSA) in leukemia, breast cancer, and prostate cancer. We constructed a TSPPI multilayer network and obtained candidate drug-target modules from the network. Gene ontology analysis provided insights into the significance of the drug-target modules and co-expression networks. Finally, two modules were obtained as potential treatment patterns for TSA. Through analysis of the significance, composition, and functions of the selected drug-target modules, we validated the feasibility and rationality of our proposed SDTP method for identifying drug treatment patterns. In summary, our novel approach used a multilayer network model to overcome the shortcomings of single-layer networks and combined the network with information on drug activity. Based on the discovered drug treatment patterns, we can predict the potential diseases that the drug can treat. That is, if a disease-related protein module has a similar structure, then the drug is likely to be a potential drug for the treatment of the disease.

Complete Topological Mapping of a Cellular Protein Interactome Reveals Bow-Tie Motifs as Ubiquitous Connectors of Protein Complexes

The network topology of a protein interactome is shaped by the function of each protein, making it a resource of functional knowledge in tissues and in single cells. Today, this resource is underused, as complete network topology characterization has proved difficult for large protein interactomes. We apply a matrix visualization and decoding approach to a physical protein interactome of a dendritic cell, thereby characterizing its topology with no prior assumptions of structure. We discover 294 proteins, each forming topological motifs called "bow-ties" that tie together the majority of observed protein complexes. The central proteins of these bow-ties have unique network properties, display multifunctional capabilities, are enriched for essential proteins, and are widely expressed in other cells and tissues. Collectively, the bow-tie motifs are a pervasive and previously unnoted topological trend in cellular interactomes. As such, these results provide fundamental knowledge on how intracellular protein connectivity is organized and operates.

ResponseNet v.3: revealing signaling and regulatory pathways connecting your proteins and genes across human tissues

ResponseNet v.3 is an enhanced version of ResponseNet, a web server that is designed to highlight signaling and regulatory pathways connecting user-defined proteins and genes by using the ResponseNet network optimization approach (http://netbio.bgu.ac.il/respnet). Users run ResponseNet by defining source and target sets of proteins, genes and/or microRNAs, and by specifying a molecular interaction network (interactome). The output of ResponseNet is a sparse, high-probability interactome subnetwork that connects the two sets, thereby revealing additional molecules and interactions that are involved in the studied condition. In recent years, massive efforts were invested in profiling the transcriptomes of human tissues, enabling the inference of human tissue interactomes. ResponseNet v.3 expands ResponseNet2.0 by harnessing ∼11,600 RNA-sequenced human tissue profiles made available by the Genotype-Tissue Expression consortium, to support context-specific analysis of 44 human tissues. Thus, ResponseNet v.3 allows users to illuminate the signaling and regulatory pathways potentially active in the context of a specific tissue, and to compare them with active pathways in other tissues. In the era of precision medicine, such analyses open the door for tissue- and patient-specific analyses of pathways and diseases.

Differential network analysis of multiple human tissue interactomes highlights tissue-selective processes and genetic disorder genes

Motivation

Differential network analysis, designed to highlight network changes between conditions, is an important paradigm in network biology. However, differential network analysis methods have been typically designed to compare between two conditions and were rarely applied to multiple protein interaction networks (interactomes). Importantly, large-scale benchmarks for their evaluation have been lacking.

Results

Here, we present a framework for assessing the ability of differential network analysis of multiple human tissue interactomes to highlight tissue-selective processes and disorders. For this, we created a benchmark of 6499 curated tissue-specific Gene Ontology biological processes. We applied five methods, including four differential network analysis methods, to construct weighted interactomes for 34 tissues. Rigorous assessment of this benchmark revealed that differential analysis methods perform well in revealing tissue-selective processes (AUCs of 0.82–0.9). Next, we applied differential network analysis to illuminate the genes underlying tissue-selective hereditary disorders. For this, we curated a dataset of 1305 tissue-specific hereditary disorders and their manifesting tissues. Focusing on subnetworks containing the top 1% differential interactions in disease-relevant tissue interactomes revealed significant enrichment for disorder-causing genes in 18.6% of the cases, with a significantly high success rate for blood, nerve, muscle and heart diseases.

Summary

Altogether, we offer a framework that includes expansive manually curated datasets of tissue-selective processes and disorders to be used as benchmarks or to illuminate tissue-selective processes and genes. Our results demonstrate that differential analysis of multiple human tissue interactomes is a powerful tool for highlighting processes and genes with tissue-selective functionality and clinical impact.

Availability and implementation

Datasets are available as part of the Supplementary data.

Supplementary information

Supplementary data are available at Bioinformatics online.

Integrated querying and version control of context-specific biological networks

MOTIVATION

Biomolecular data stored in public databases is increasingly specialized to organisms, context/pathology and tissue type, potentially resulting in significant overhead for analyses. These networks are often specializations of generic interaction sets, presenting opportunities for reducing storage and computational cost. Therefore, it is desirable to develop effective compression and storage techniques, along with efficient algorithms and a flexible query interface capable of operating on compressed data structures. Current graph databases offer varying levels of support for network integration. However, these solutions do not provide efficient methods for the storage and querying of versioned networks.

RESULTS

We present VerTIoN, a framework consisting of novel data structures and associated query mechanisms for integrated querying of versioned context-specific biological networks. As a use case for our framework, we study network proximity queries in which the user can select and compose a combination of tissue-specific and generic networks. Using our compressed version tree data structure, in conjunction with state-of-the-art numerical techniques, we demonstrate real-time querying of large network databases.

CONCLUSION

Our results show that it is possible to support flexible queries defined on heterogeneous networks composed at query time while drastically reducing response time for multiple simultaneous queries. The flexibility offered by VerTIoN in composing integrated network versions opens significant new avenues for the utilization of ever increasing volume of context-specific network data in a broad range of biomedical applications.

AVAILABILITY AND IMPLEMENTATION

VerTIoN is implemented as a C++ library and is available at http://compbio.case.edu/omics/software/vertion and https://github.com/tjcowman/vertion.

CONTACT

tyler.cowman@case.edu.

Reconstruction of Cell-type-Specific Interactomes at Single-Cell Resolution

The human interactome is instrumental in the systems-level study of the cell and the contextualization of disease-associated gene perturbations. However, reference organismal interactomes do not capture the cell-type-specific context in which proteins and modules preferentially act. Here, we introduce SCINET, a computational framework that reconstructs an ensemble of cell-type-specific interactomes by integrating a global, context-independent reference interactome with a single-cell gene-expression profile. SCINET addresses technical challenges of single-cell data by robustly imputing, transforming, and normalizing the initially noisy and sparse expression of data. Inferred cell-level gene interaction probabilities and group-level interaction strengths define cell-type-specific interactomes. We use SCINET to reconstruct and analyze interactomes of the major human brain and immune cell types, revealing specificity and modularity of perturbations associated with neurodegenerative, neuropsychiatric, and autoimmune disorders. We report cell-type interactomes for brain and immune cell types, together with the SCINET package.

Biological Network Approaches and Applications in Rare Disease Studies

Network biology has the capability to integrate, represent, interpret, and model complex biological systems by collectively accommodating biological omics data, biological interactions and associations, graph theory, statistical measures, and visualizations. Biological networks have recently been shown to be very useful for studies that decipher biological mechanisms and disease etiologies and for studies that predict therapeutic responses, at both the molecular and system levels. In this review, we briefly summarize the general framework of biological network studies, including data resources, network construction methods, statistical measures, network topological properties, and visualization tools. We also introduce several recent biological network applications and methods for the studies of rare diseases.

Bi-level feature selection in high dimensional AFT models with applications to a genomic study

We propose a new bi-level feature selection method for high dimensional accelerated failure time models by formulating the models to a single index model. The method yields sparse solutions at both the group and individual feature levels along with an expedient algorithm, which is computationally efficient and easily implemented. We analyze a genomic dataset for an illustration, and present a simulation study to show the finite sample performance of the proposed method.

SoyCSN: Soybean context-specific network analysis and prediction based on tissue-specific transcriptome data

The Soybean Gene Atlas project provides a comprehensive map for understanding gene expression patterns in major soybean tissues from flower, root, leaf, nodule, seed, and shoot and stem. The RNA-Seq data generated in the project serve as a valuable resource for discovering tissue-specific transcriptome behavior of soybean genes in different tissues. We developed a computational pipeline for Soybean context-specific network (SoyCSN) inference with a suite of prediction tools to analyze, annotate, retrieve, and visualize soybean context-specific networks at both transcriptome and interactome levels. BicMix and Cross-Conditions Cluster Detection algorithms were applied to detect modules based on co-expression relationships across all the tissues. Soybean context-specific interactomes were predicted by combining soybean tissue gene expression and protein-protein interaction data. Functional analyses of these predicted networks provide insights into soybean tissue specificities. For example, under symbiotic, nitrogen-fixing conditions, the constructed soybean leaf network highlights the connection between the photosynthesis function and rhizobium-legume symbiosis. SoyCSN data and all its results are publicly available via an interactive web service within the Soybean Knowledge Base (SoyKB) at http://soykb.org/SoyCSN. SoyCSN provides a useful web-based access for exploring context specificities systematically in gene regulatory mechanisms and gene relationships for soybean researchers and molecular breeders.

Tagger-A Swiss army knife for multiomics to dissect cell type-specific mechanisms of gene expression in mice

A deep understanding of how regulation of the multiple levels of gene expression in mammalian tissues give rise to complex phenotypes has been impeded by cellular diversity. A handful of techniques were developed to tag-select nucleic acids of interest in specific cell types, thereby enabling their capture. We expanded this strategy by developing the Tagger knock-in mouse line bearing a quad-cistronic transgene combining enrichment tools for nuclei, nascent RNA, translating mRNA, and mature microRNA (miRNA). We demonstrate that Tagger can capture the desired nucleic acids, enabling multiple omics approaches to be applied to specific cell types in vivo using a single transgenic mouse line.

This Methods and Resources paper describes Tagger, a knock-in mouse line bearing a quad-cistronic transgene that enables the capture of translating mRNAs, mature miRNAs, pulse-labeled total RNA, and the nucleus, all from specific cells of complex tissues.

Systematic genetic analyses of GWAS data reveal an association between the immune system and insomnia

BACKGROUND

Previous studies have inferred a strong genetic component for insomnia. However, the etiology of insomnia is still unclear. The aim of the current study was to explore potential biological pathways, gene networks, and brain regions associated with insomnia.

METHODS

Using pathways (gene sets) from Reactome, we carried out a two-stage gene set enrichment analysis strategy. From a large genome-wide association studies (GWASs) of insomnia symptoms (32,155 cases/26,973 controls), significant gene sets were tested for replication in other large GWASs of insomnia complaints (32,384 cases/80,622 controls). After the network analysis of unique genes within the replicated pathways, a gene set analysis for genes in each cluster/module of the enhancing neuroimaging genetics through meta-analysis GWAS data was performed for the volumes of the intracranial and seven subcortical regions.

RESULTS

A total of 31 of 1,816 Reactome pathways were identified and showed associations with insomnia risk. In addition, seven functionally and topologically interconnected clusters (clusters 0-6) and six gene modules (named Yellow, Blue, Brown, Green, Red, and Turquoise) were associated with insomnia. Moreover, significant associations were detected between common variants of the genes in Cluster 2 with hippocampal volume (p = 0.035; family wise error [FWE] correction) and the red module with intracranial volume (p = 0.047; FWE correction). Functional enrichment for genes in the Cluster 2 and the Red module revealed the involvement of immune responses, nervous system development, NIK/NF-kappaB signaling, and I-kappaB kinase/NF-kappaB signaling. Core genes (UBC, UBB, and UBA52) in the interconnected functional network were found to be involved in regulating brain development.

CONCLUSIONS

The current study demonstrates that the immune system and the hippocampus may play central roles in neurodevelopment and insomnia risk.

Disease classification: from phenotypic similarity to integrative genomics and beyond

A fundamental challenge of modern biomedical research is understanding how diseases that are similar on the phenotypic level are similar on the molecular level. Integration of various genomic data sets with the traditionally used phenotypic disease similarity revealed novel genetic and molecular mechanisms and blurred the distinction between monogenic (Mendelian) and complex diseases. Network-based medicine has emerged as a complementary approach for identifying disease-causing genes, genetic mediators, disruptions in the underlying cellular functions and for drug repositioning. The recent development of machine and deep learning methods allow for leveraging real-life information about diseases to refine genetic and phenotypic disease relationships. This review describes the historical development and recent methodological advancements for studying disease classification (nosology).

Molecular Signatures of Fusion Proteins in Cancer

Although gene fusions are recognized as driver mutations in a wide variety of cancers, the general molecular mechanisms underlying oncogenic fusion proteins are insufficiently understood. Here, we employ large-scale data integration and machine learning and (1) identify three functionally distinct subgroups of gene fusions and their molecular signatures; (2) characterize the cellular pathways rewired by fusion events across different cancers; and (3) analyze the relative importance of over 100 structural, functional, and regulatory features of ∼2200 gene fusions. We report subgroups of fusions that likely act as driver mutations and find that gene fusions disproportionately affect pathways regulating cellular shape and movement. Although fusion proteins are similar across different cancer types, they affect cancer type-specific pathways. Key indicators of fusion-forming proteins include high and nontissue specific expression, numerous splice sites, and higher centrality in protein-interaction networks. Together, these findings provide unifying and cancer type-specific trends across diverse oncogenic fusion proteins.

Collective influencers in protein interaction networks

Recent research increasingly shows the relevance of network based approaches for our understanding of biological systems. Analyzing human protein interaction networks, we determined collective influencers (CI), defined as network nodes that damage the integrity of the underlying networks to the utmost degree. We found that CI proteins were enriched with essential, regulatory, signaling and disease genes as well as drug targets, indicating their biological significance. Also by focusing on different organisms, we found that CI proteins had a penchant to be evolutionarily conserved as CI proteins, indicating the fundamental role that collective influencers in protein interaction networks plays for our understanding of regulation, diseases and evolution.

IID 2018 update: context-specific physical protein-protein interactions in human, model organisms and domesticated species

Knowing the set of physical protein-protein interactions (PPIs) that occur in a particular context-a tissue, disease, or other condition-can provide valuable insights into key research questions. However, while the number of identified human PPIs is expanding rapidly, context information remains limited, and for most non-human species context-specific networks are completely unavailable. The Integrated Interactions Database (IID) provides one of the most comprehensive sets of context-specific human PPI networks, including networks for 133 tissues, 91 disease conditions, and many other contexts. Importantly, it also provides context-specific networks for 17 non-human species including model organisms and domesticated animals. These species are vitally important for drug discovery and agriculture. IID integrates interactions from multiple databases and datasets. It comprises over 4.8 million PPIs annotated with several types of context: tissues, subcellular localizations, diseases, and druggability information (the latter three are new annotations not available in the previous version). This update increases the number of species from 6 to 18, the number of PPIs from ∼1.5 million to ∼4.8 million, and the number of tissues from 30 to 133. IID also now supports topology and enrichment analyses of returned networks. IID is available at http://ophid.utoronto.ca/iid.

A prior-based integrative framework for functional transcriptional regulatory network inference

Transcriptional regulatory networks specify regulatory proteins controlling the context-specific expression levels of genes. Inference of genome-wide regulatory networks is central to understanding gene regulation, but remains an open challenge. Expression-based network inference is among the most popular methods to infer regulatory networks, however, networks inferred from such methods have low overlap with experimentally derived (e.g. ChIP-chip and transcription factor (TF) knockouts) networks. Currently we have a limited understanding of this discrepancy. To address this gap, we first develop a regulatory network inference algorithm, based on probabilistic graphical models, to integrate expression with auxiliary datasets supporting a regulatory edge. Second, we comprehensively analyze our and other state-of-the-art methods on different expression perturbation datasets. Networks inferred by integrating sequence-specific motifs with expression have substantially greater agreement with experimentally derived networks, while remaining more predictive of expression than motif-based networks. Our analysis suggests natural genetic variation as the most informative perturbation for network inference, and, identifies core TFs whose targets are predictable from expression. Multiple reasons make the identification of targets of other TFs difficult, including network architecture and insufficient variation of TF mRNA level. Finally, we demonstrate the utility of our inference algorithm to infer stress-specific regulatory networks and for regulator prioritization.

Predicting multicellular function through multi-layer tissue networks

Motivation

Understanding functions of proteins in specific human tissues is essential for insights into disease diagnostics and therapeutics, yet prediction of tissue-specific cellular function remains a critical challenge for biomedicine.

Results

Here, we present OhmNet, a hierarchy-aware unsupervised node feature learning approach for multi-layer networks. We build a multi-layer network, where each layer represents molecular interactions in a different human tissue. OhmNet then automatically learns a mapping of proteins, represented as nodes, to a neural embedding-based low-dimensional space of features. OhmNet encourages sharing of similar features among proteins with similar network neighborhoods and among proteins activated in similar tissues. The algorithm generalizes prior work, which generally ignores relationships between tissues, by modeling tissue organization with a rich multiscale tissue hierarchy. We use OhmNet to study multicellular function in a multi-layer protein interaction network of 107 human tissues. In 48 tissues with known tissue-specific cellular functions, OhmNet provides more accurate predictions of cellular function than alternative approaches, and also generates more accurate hypotheses about tissue-specific protein actions. We show that taking into account the tissue hierarchy leads to improved predictive power. Remarkably, we also demonstrate that it is possible to leverage the tissue hierarchy in order to effectively transfer cellular functions to a functionally uncharacterized tissue. Overall, OhmNet moves from flat networks to multiscale models able to predict a range of phenotypes spanning cellular subsystems.

Availability and implementation

Source code and datasets are available at http://snap.stanford.edu/ohmnet.

Systematic interrogation of diverse Omic data reveals interpretable, robust, and generalizable transcriptomic features of clinically successful therapeutic targets

Target selection is the first and pivotal step in drug discovery. An incorrect choice may not manifest itself for many years after hundreds of millions of research dollars have been spent. We collected a set of 332 targets that succeeded or failed in phase III clinical trials, and explored whether Omic features describing the target genes could predict clinical success. We obtained features from the recently published comprehensive resource: Harmonizome. Nineteen features appeared to be significantly correlated with phase III clinical trial outcomes, but only 4 passed validation schemes that used bootstrapping or modified permutation tests to assess feature robustness and generalizability while accounting for target class selection bias. We also used classifiers to perform multivariate feature selection and found that classifiers with a single feature performed as well in cross-validation as classifiers with more features (AUROC = 0.57 and AUPR = 0.81). The two predominantly selected features were mean mRNA expression across tissues and standard deviation of expression across tissues, where successful targets tended to have lower mean expression and higher expression variance than failed targets. This finding supports the conventional wisdom that it is favorable for a target to be present in the tissue(s) affected by a disease and absent from other tissues. Overall, our results suggest that it is feasible to construct a model integrating interpretable target features to inform target selection. We anticipate deeper insights and better models in the future, as researchers can reuse the data we have provided to improve methods for handling sample biases and learn more informative features. Code, documentation, and data for this study have been deposited on GitHub at https://github.com/arouillard/omic-features-successful-targets.

Another Round of "Clue" to Uncover the Mystery of Complex Traits

A plethora of genetic association analyses have identified several genetic risk loci. Technological and statistical advancements have now led to the identification of not only common genetic variants, but also low-frequency variants, structural variants, and environmental factors, as well as multi-omics variations that affect the phenotypic variance of complex traits in a population, thus referred to as complex trait architecture. The concept of heritability, or the proportion of phenotypic variance due to genetic inheritance, has been studied for several decades, but its application is mainly in addressing the narrow sense heritability (or additive genetic component) from Genome-Wide Association Studies (GWAS). In this commentary, we reflect on our perspective on the complexity of understanding heritability for human traits in comparison to model organisms, highlighting another round of clues beyond GWAS and an alternative approach, investigating these clues comprehensively to help in elucidating the genetic architecture of complex traits.