Searching the overlap between network modules with specific betweeness (S2B) and its application to cross-disease analysis

Open MoA: revealing the mechanism of action (MoA) based on network topology and hierarchy

MOTIVATION

Many approaches in systems biology have been applied in drug repositioning due to the increased availability of the omics data and computational biology tools. Using a multi-omics integrated network, which contains information of various biological interactions, could offer a more comprehensive inspective and interpretation for the drug mechanism of action (MoA).

RESULTS

We developed a computational pipeline for dissecting the hidden MoAs of drugs (Open MoA). Our pipeline computes confidence scores to edges that represent connections between genes/proteins in the integrated network. The interactions showing the highest confidence score could indicate potential drug targets and infer the underlying molecular MoAs. Open MoA was also validated by testing some well-established targets. Additionally, we applied Open MoA to reveal the MoA of a repositioned drug (JNK-IN-5A) that modulates the PKLR expression in HepG2 cells and found STAT1 is the key transcription factor. Overall, Open MoA represents a first-generation tool that could be utilized for predicting the potential MoA of repurposed drugs and dissecting de novo targets for developing effective treatments.

AVAILABILITY AND IMPLEMENTATION

Source code is available at https://github.com/XinmengLiao/Open_MoA.

Topology Driven Analysis of Protein - Protein Interactome for Prioritizing Key Comorbid Genes via Sub Graph Based Average Path Length Centrality

In gene-based therapies, local perturbations associated with one disease can lead to comorbidity as it influences the pathways involved with the other diseases. The key genes orchestrating the common biological mechanisms are need to be prioritized for addressing the challenges introduced by the cross talks between disease modules. Here, a local centrality measure named Sub graph based Average Path length Double Specific Betweenness centrality (SAPDSB) for prioritizing the comorbid genes via Protein-Protein Interaction Network (PPIN) analysis is presented. This approach can be used to identify putative biomarkers which can be repurposed for the management of comorbidity. Proposed network based topological measure is designed specifically to prioritize the comorbid genes that are most likely to be present in the overlap of disease modules. In order to attain this, the estimated average path length of the seed network which holds Protein-Protein Interactions (PPIs) of the disease genes is exploited. Prioritized comorbid genes are further pruned using centrality-based cut-off values and specificity scores. The biological significance of the resultant genes is corroborated with connectivity analysis using leave-one-out method, pathway enrichment analysis and a comparative analysis using single disease-based gene prioritization tools. For performance analysis, proposed approach is tested using case studies involving common diseases and rare neurodegenerative diseases. For case study1, diseases such as Diabetes, Carcinoma and Alzheimer's are considered in a pairwise manner while for case study2, Amyotrophic Lateral Sclerosis (ALS) and Spinal Muscular Atrophy (SMA) are considered. As outcome, prioritized candidate genes and biological pathways associated with respective disease pairs have been found. The associations from top 10 candidate genes in different disease pair combinations of Diabetes-Carcinoma-Alzheimer's revealed common genes like CREBBP, TP53, HSP90AA1 and the common pathway namely p53 pathway feedback loops 2. Out of the pathways retrieved from the top 10 genes associated with ALS-SMA disease pair, 60% of unique pathways are found to be leading to both diseases and its comorbidities. Comparative analysis of the proposed method with recent similar approach also reported a clear degree of benefits in performance.

Biological networks and complexity in early-onset motor neuron diseases

Motor neuron diseases (MNDs) are neuromuscular disorders where the spinal motor neurons-either the cell bodies themselves or their axons-are the primary cells affected. To date, there are 120 different genes that are lost or mutated in pediatric-onset MNDs. Most of these childhood-onset disorders, aside from spinal muscular atrophy (SMA), lack viable therapeutic options. Previous research on MNDs has focused on understanding the pathobiology of a single, specific gene mutation and targeting therapies to that pathobiology. This reductionist approach has yielded therapeutic options for a specific disorder, in this case SMA. Unfortunately, therapies specific for SMA have not been effective against other pediatric-onset MNDs. Pursuing the same approach for the other defined MNDs would require development of at least 120 independent treatments raising feasibility issues. We propose an alternative to this this type of reductionist approach by conceptualizing MNDs in a complex adaptive systems framework that will allow identification of common molecular and cellular pathways which form biological networks that are adversely affected in early-onset MNDs and thus MNDs with similar phenotypes despite diverse genotypes. This systems biology approach highlights the complexity and self-organization of the motor system as well as the ways in which it can be affected by these genetic disorders. Using this integrated approach to understand early-onset MNDs, we would be better poised to expand the therapeutic repertoire for multiple MNDs.

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.

MOSES: A New Approach to Integrate Interactome Topology and Functional Features for Disease Gene Prediction

Disease gene prediction is to date one of the main computational challenges of precision medicine. It is still uncertain if disease genes have unique functional properties that distinguish them from other non-disease genes or, from a network perspective, if they are located randomly in the interactome or show specific patterns in the network topology. In this study, we propose a new method for disease gene prediction based on the use of biological knowledge-bases (gene-disease associations, genes functional annotations, etc.) and interactome network topology. The proposed algorithm called MOSES is based on the definition of two somewhat opposing sets of genes both disease-specific from different perspectives: warm seeds (i.e., disease genes obtained from databases) and cold seeds (genes far from the disease genes on the interactome and not involved in their biological functions). The application of MOSES to a set of 40 diseases showed that the suggested putative disease genes are significantly enriched in their reference disease. Reassuringly, known and predicted disease genes together, tend to form a connected network module on the human interactome, mitigating the scattered distribution of disease genes which is probably due to both the paucity of disease-gene associations and the incompleteness of the interactome.

Optimal Bayesian supervised domain adaptation for RNA sequencing data

MOTIVATION

When learning to subtype complex disease based on next-generation sequencing data, the amount of available data is often limited. Recent works have tried to leverage data from other domains to design better predictors in the target domain of interest with varying degrees of success. But they are either limited to the cases requiring the outcome label correspondence across domains or cannot leverage the label information at all. Moreover, the existing methods cannot usually benefit from other information available a priori such as gene interaction networks.

RESULTS

In this paper, we develop a generative optimal Bayesian supervised domain adaptation (OBSDA) model that can integrate RNA sequencing (RNA-Seq) data from different domains along with their labels for improving prediction accuracy in the target domain. Our model can be applied in cases where different domains share the same labels or have different ones. OBSDA is based on a hierarchical Bayesian negative binomial model with parameter factorization, for which the optimal predictor can be derived by marginalization of likelihood over the posterior of the parameters. We first provide an efficient Gibbs sampler for parameter inference in OBSDA. Then, we leverage the gene-gene network prior information and construct an informed and flexible variational family to infer the posterior distributions of model parameters. Comprehensive experiments on real-world RNA-Seq data demonstrate the superior performance of OBSDA, in terms of accuracy in identifying cancer subtypes by utilizing data from different domains. Moreover, we show that by taking advantage of the prior network information we can further improve the performance.

AVAILABILITY

The source code for implementations of OBSDA and SI-OBSDA are available at the following link. https://github.com/SHBLK/BSDA.

MODifieR: an Ensemble R Package for Inference of Disease Modules from Transcriptomics Networks

MOTIVATION

Complex diseases are due to the dense interactions of many disease-associated factors that dysregulate genes that in turn form the so-called disease modules, which have shown to be a powerful concept for understanding pathological mechanisms. There exist many disease module inference methods that rely on somewhat different assumptions, but there is still no gold standard or best-performing method. Hence, there is a need for combining these methods to generate robust disease modules.

RESULTS

We developed MODule IdentiFIER (MODifieR), an ensemble R package of nine disease module inference methods from transcriptomics networks. MODifieR uses standardized input and output allowing the possibility to combine individual modules generated from these methods into more robust disease-specific modules, contributing to a better understanding of complex diseases.

AVAILABILITY AND IMPLEMENTATION

MODifieR is available under the GNU GPL license and can be freely downloaded from https://gitlab.com/Gustafsson-lab/MODifieR and as a Docker image from https://hub.docker.com/r/ddeweerd/modifier.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Discovering the genes mediating the interactions between chronic respiratory diseases in the human interactome

The molecular and clinical features of a complex disease can be influenced by other diseases affecting the same individual. Understanding disease-disease interactions is therefore crucial for revealing shared molecular mechanisms among diseases and designing effective treatments. Here we introduce Flow Centrality (FC), a network-based approach to identify the genes mediating the interaction between two diseases in a protein-protein interaction network. We focus on asthma and COPD, two chronic respiratory diseases that have been long hypothesized to share common genetic determinants and mechanisms. We show that FC highlights potential mediator genes between the two diseases, and observe similar outcomes when applying FC to 66 additional pairs of related diseases. Further, we perform in vitro perturbation experiments on a widely replicated asthma gene, GSDMB, showing that FC identifies candidate mediators of the interactions between GSDMB and COPD-associated genes. Our results indicate that FC predicts promising gene candidates for further study of disease-disease interactions.

Network Biology Identifies Novel Regulators of CFTR Trafficking and Membrane Stability

In cystic fibrosis, the most common disease-causing mutation is F508del, which causes not only intracellular retention and degradation of CFTR, but also defective channel gating and decreased membrane stability of the small amount that reaches the plasma membrane (PM). Thus, pharmacological correction of mutant CFTR requires targeting of multiple cellular defects in order to achieve clinical benefit. Although small-molecule compounds have been identified and commercialized that can correct its folding or gating, an efficient retention of F508del CFTR at the PM has not yet been explored pharmacologically despite being recognized as a crucial factor for improving functional rescue of chloride transport. In ongoing efforts to determine the CFTR interactome at the PM, we used three complementary approaches: targeting proteins binding to tyrosine-phosphorylated CFTR, protein complexes involved in cAMP-mediated CFTR stabilization at the PM, and proteins selectively interacting at the PM with rescued F508del-CFTR but not wt-CFTR. Using co-immunoprecipitation or peptide–pull down strategies, we identified around 400 candidate proteins through sequencing of complex protein mixtures using the nano-LC Triple TOF MS technique. Key candidate proteins were validated for their robust interaction with CFTR-containing protein complexes and for their ability to modulate the amount of CFTR expressed at the cell surface of bronchial epithelial cells. Here, we describe how we explored the abovementioned experimental datasets to build a protein interaction network with the aim of identifying novel pharmacological targets to rescue CFTR function in cystic fibrosis (CF) patients. We identified and validated novel candidate proteins that were essential components of the network but not detected in previous proteomic analyses.