DSEATM: drug set enrichment analysis uncovering disease mechanisms by biomedical text mining

Disease pathogenesis is always a major topic in biomedical research. With the exponential growth of biomedical information, drug effect analysis for specific phenotypes has shown great promise in uncovering disease-associated pathways. However, this method has only been applied to a limited number of drugs. Here, we extracted the data of 4634 diseases, 3671 drugs, 112 809 disease-drug associations and 81 527 drug-gene associations by text mining of 29 168 919 publications. On this basis, we proposed a 'Drug Set Enrichment Analysis by Text Mining (DSEATM)' pipeline and applied it to 3250 diseases, which outperformed the state-of-the-art method. Furthermore, diseases pathways enriched by DSEATM were similar to those obtained using the TCGA cancer RNA-seq differentially expressed genes. In addition, the drug number, which showed a remarkable positive correlation of 0.73 with the AUC, plays a determining role in the performance of DSEATM. Taken together, DSEATM is an auspicious and accurate disease research tool that offers fresh insights.

Integrative transcriptome-wide analysis of atopic dermatitis for drug repositioning

Atopic dermatitis (AD) is one of the most common inflammatory skin diseases, which significantly impact the quality of life. Transcriptome-wide association study (TWAS) was conducted to estimate both transcriptomic and genomic features of AD and detected significant associations between 31 expression quantitative loci and 25 genes. Our results replicated well-known genetic markers for AD, as well as 4 novel associated genes. Next, transcriptome meta-analysis was conducted with 5 studies retrieved from public databases and identified 5 additional novel susceptibility genes for AD. Applying the connectivity map to the results from TWAS and meta-analysis, robustly enriched perturbations were identified and their chemical or functional properties were analyzed. Here, we report the first research on integrative approaches for an AD, combining TWAS and transcriptome meta-analysis. Together, our findings could provide a comprehensive understanding of the pathophysiologic mechanisms of AD and suggest potential drug candidates as alternative treatment options.

Integrative genomic and transcriptomic analyses on publicly available data-sets together with in silico drug repositioning identifies alternative therapeutic options to treat atopic dermatitis.

Agent Clustering Strategy Based on Metabolic Flux Distribution and Transcriptome Expression for Novel Drug Development

The network module-based method has been used for drug repositioning. The traditional drug repositioning method only uses the gene characteristics of the drug but ignores the drug-triggered metabolic changes. The metabolic network systematically characterizes the connection between genes, proteins, and metabolic reactions. The differential metabolic flux distribution, as drug metabolism characteristics, was employed to cluster the agents with similar MoAs (mechanism of action). In this study, agents with the same pharmacology were clustered into one group, and a total of 1309 agents from the CMap database were clustered into 98 groups based on differential metabolic flux distribution. Transcription factor (TF) enrichment analysis revealed the agents in the same group (such as group 7 and group 26) were confirmed to have similar MoAs. Through this agent clustering strategy, the candidate drugs which can inhibit (Japanese encephalitis virus) JEV infection were identified. This study provides new insights into drug repositioning and their MoAs.

MNBDR: A Module Network Based Method for Drug Repositioning

Drug repurposing/repositioning, which aims to find novel indications for existing drugs, contributes to reducing the time and cost for drug development. For the recent decade, gene expression profiles of drug stimulating samples have been successfully used in drug repurposing. However, most of the existing methods neglect the gene modules and the interactions among the modules, although the cross-talks among pathways are common in drug response. It is essential to develop a method that utilizes the cross-talks information to predict the reliable candidate associations. In this study, we developed MNBDR (Module Network Based Drug Repositioning), a novel method that based on module network to screen drugs. It integrated protein-protein interactions and gene expression profile of human, to predict drug candidates for diseases. Specifically, the MNBDR mined dense modules through protein-protein interaction (PPI) network and constructed a module network to reveal cross-talks among modules. Then, together with the module network, based on existing gene expression data set of drug stimulation samples and disease samples, we used random walk algorithms to capture essential modules in disease development and proposed a new indicator to screen potential drugs for a given disease. Results showed MNBDR could provide better performance than popular methods. Moreover, functional analysis of the essential modules in the network indicated our method could reveal biological mechanism in drug response.

A Machine Learning Method for Drug Combination Prediction

Drug combination is now a hot research topic in the pharmaceutical industry, but experiment-based methodologies are extremely costly in time and money. Many computational methods have been proposed to address these problems by starting from existing drug combinations. However, in most cases, only molecular structure information is included, which covers too limited a set of drug characteristics to efficiently screen drug combinations. Here, we integrated similarity-based multifeature drug data to improve the prediction accuracy by using the neighbor recommender method combined with ensemble learning algorithms. By conducting feature assessment analysis, we selected the most useful drug features and achieved 0.964 AUC in the ensemble models. The comparison results showed that the ensemble models outperform traditional machine learning algorithms such as support vector machine (SVM), naïve Bayes (NB), and logistic regression (GLM). Furthermore, we predicted 7 candidate drug combinations for a specific drug, paclitaxel, and successfully verified that the two of the predicted combinations have promising effects.

Heat Diffusion Kernel Algorithm-Based Interpretation of the Disease Intervention Mechanism for DHA

Docosahexaenoic acid (DHA) is effective in the prevention and treatment of cancer, congenital disorders, and various chronic diseases. According to the omnigenic hypothesis, these complex diseases are caused by disordered gene regulatory networks comprising dozens to hundreds of core genes and a mass of peripheral genes. However, conventional research on the disease intervention mechanism of DHA only focused on specific types of genes or pathways instead of examining genes at the network level, resulting in conflicting conclusions. In this study, we used HotNet2, a heat diffusion kernel algorithm, to calculate the gene regulatory networks of connectivity map (cMap)-derived agents (including DHA) based on gene expression profiles, aiming to interpret the disease intervention mechanism of DHA at the network level. As a result, significant gene regulatory networks for DHA and 676 cMap-derived agents were identified respectively. The biological functions of the DHA-regulated gene network provide preliminary insights into the mechanism by which DHA intervenes in disease. In addition, we compared the gene regulatory networks of DHA with those of cMap-derived agents, which allowed us to predict the pharmacological effects and disease intervention mechanism of DHA by analogy with similar agents with clear indications and mechanisms. Some of our analysis results were supported by experimental observations. Therefore, this study makes a significant contribution to research on the disease intervention mechanism of DHA at the regulatory network level, demonstrating the potential application value of this methodology in clarifying the mechanisms about nutrients influencing health.

It is time to apply biclustering: a comprehensive review of biclustering applications in biological and biomedical data

Biclustering is a powerful data mining technique that allows clustering of rows and columns, simultaneously, in a matrix-format data set. It was first applied to gene expression data in 2000, aiming to identify co-expressed genes under a subset of all the conditions/samples. During the past 17 years, tens of biclustering algorithms and tools have been developed to enhance the ability to make sense out of large data sets generated in the wake of high-throughput omics technologies. These algorithms and tools have been applied to a wide variety of data types, including but not limited to, genomes, transcriptomes, exomes, epigenomes, phenomes and pharmacogenomes. However, there is still a considerable gap between biclustering methodology development and comprehensive data interpretation, mainly because of the lack of knowledge for the selection of appropriate biclustering tools and further supporting computational techniques in specific studies. Here, we first deliver a brief introduction to the existing biclustering algorithms and tools in public domain, and then systematically summarize the basic applications of biclustering for biological data and more advanced applications of biclustering for biomedical data. This review will assist researchers to effectively analyze their big data and generate valuable biological knowledge and novel insights with higher efficiency.

Identification of NUDT5 Inhibitors From Approved Drugs

Recent studies have revealed the important role of NUDT5 in estrogen signaling and breast cancer, but research on the corresponding targeted therapy has just started. Drug repositioning strategy can effectively reduce the time and economic resources spent on drug discovery. To find novel inhibitors of NUDT5, we investigated the previously identified connectivity map-based drug association models and found eighteen FDA approved drugs as candidates. The molecular docking and molecular dynamic simulation were performed and revealed that fourteen organic drugs have the potential to bind the NUDT5 target. Eight representative drugs were selected to perform the cell line viability inhibition analysis, and the results showed that seven of them were able to suppress MCF7 breast cancer cells. Two drugs, nomifensine and isoconazole, showed lower IC₅₀ than the known antiestrogens raloxifene and tamoxifen, and they deserve further pharmacodynamic investigations to test their feasibility for use as NUDT5 inhibitors.

Systems Chemical Genetics-Based Drug Discovery: Prioritizing Agents Targeting Multiple/Reliable Disease-Associated Genes as Drug Candidates

Genetic disease genes are considered a promising source of drug targets. Most diseases are caused by more than one pathogenic factor; thus, it is reasonable to consider that chemical agents targeting multiple disease genes are more likely to have desired activities. This is supported by a comprehensive analysis on the relationships between agent activity/druggability and target genetic characteristics. The therapeutic potential of agents increases steadily with increasing number of targeted disease genes, and can be further enhanced by strengthened genetic links between targets and diseases. By using the multi-label classification models for genetics-based drug activity prediction, we provide universal tools for prioritizing drug candidates. All of the documented data and the machine-learning prediction service are available at SCG-Drug (http://zhanglab.hzau.edu.cn/scgdrug).

Dose-dependent target diversion of Danhong injection on the Glu-GLT-1/Gly-GlyRα dynamic balance module of cerebral ischemia

Function-oriented modular structure analysis is a great challenge in module-based pharmacological studies. A strategy to uncover target-target interaction (TTI) and dynamic balance regularity (DBR) was established to discover the structural factors influencing modular functions and explore the mechanism of Danhong injection (DHI) in treating cerebral ischemia. The dose-related metabolic features of DHI intervention were investigated using metabolomics and modular pharmacology. The findings indicated that Glu/Gly was a biomarker and Glu-GLT-1/Gly-GlyRα was the core unit regulated by DHI. Gly and Glu displayed opposite patterns and functional roles, representing intra-modular balance. GlyRα was identified as the upstream target and GLT-1 as the downstream target by inhibiting or activating GlyRα, indicating that DHI has two dose-dependent regulatory modes. GlyRα was the major target at low doses, while GLT-1 was activated as the dominant target as doses accumulated. Our study reveals that target-target interaction and dynamic balance regularity are the key factors influencing modular functions, which is a promising breakthrough for module-based pharmacological studies.

Transcription Factor NRF2 as a Therapeutic Target for Chronic Diseases: A Systems Medicine Approach

Systems medicine has a mechanism-based rather than a symptom- or organ-based approach to disease and identifies therapeutic targets in a nonhypothesis-driven manner. In this work, we apply this to transcription factor nuclear factor (erythroid-derived 2)-like 2 (NRF2) by cross-validating its position in a protein-protein interaction network (the NRF2 interactome) functionally linked to cytoprotection in low-grade stress, chronic inflammation, metabolic alterations, and reactive oxygen species formation. Multiscale network analysis of these molecular profiles suggests alterations of NRF2 expression and activity as a common mechanism in a subnetwork of diseases (the NRF2 diseasome). This network joins apparently heterogeneous phenotypes such as autoimmune, respiratory, digestive, cardiovascular, metabolic, and neurodegenerative diseases, along with cancer. Importantly, this approach matches and confirms in silico several applications for NRF2-modulating drugs validated in vivo at different phases of clinical development. Pharmacologically, their profile is as diverse as electrophilic dimethyl fumarate, synthetic triterpenoids like bardoxolone methyl and sulforaphane, protein-protein or DNA-protein interaction inhibitors, and even registered drugs such as metformin and statins, which activate NRF2 and may be repurposed for indications within the NRF2 cluster of disease phenotypes. Thus, NRF2 represents one of the first targets fully embraced by classic and systems medicine approaches to facilitate both drug development and drug repurposing by focusing on a set of disease phenotypes that appear to be mechanistically linked. The resulting NRF2 drugome may therefore rapidly advance several surprising clinical options for this subset of chronic diseases.

Structural Basis for Great Protein-Binding Potential of Flavonoids: A Case Study of Quercetin

Flavonoids exhibit a wide variety of biological effects and hold great protein-binding potential. In this study, we take quercetin, an extensively studied flavonoid, as an example to investigate the structural basis underlying their excellent protein-binding ability. Through calculating the pocket similarity of quercetin-binding proteins and comparing the binding modes between quercetins bound with different proteins, it was found that the great protein-binding capacity of quercetin not only arises from the structural conservation of target proteins, but also stems from its unique structure, in which molecular rigidity and flexibility is well-balanced and various chemical interactions could be formed between the small molecule and protein targets.

Statistical characterization of therapeutic protein modifications

Peptide mapping with liquid chromatography–tandem mass spectrometry (LC-MS/MS) is an important analytical method for characterization of post-translational and chemical modifications in therapeutic proteins. Despite its importance, there is currently no consensus on the statistical analysis of the resulting data. In this manuscript, we distinguish three statistical goals for therapeutic protein characterization: (1) estimation of site occupancy of modifications in one condition, (2) detection of differential site occupancy between conditions, and (3) estimation of combined site occupancy across multiple modification sites. We propose an approach, which addresses these goals in terms of summarizing the quantitative information from the mass spectra, statistical modeling, and model-based analysis of LC-MS/MS data. We illustrate the approach using an LC-MS/MS experiment from an antibody-drug conjugate and its monoclonal antibody intermediate. The performance was compared to a ‘naïve’ data analysis approach, by using computer simulation, evaluation of differential site occupancy in positive and negative controls, and comparisons of estimated site occupancy with orthogonal experimental measurements of N-linked glycoforms and total oxidation. The results demonstrated the importance of replicated studies of protein characterization, and of appropriate statistical modeling, for reproducible, accurate and efficient site occupancy estimation and differential analysis.

Rectified factor networks for biclustering of omics data

MOTIVATION

Biclustering has become a major tool for analyzing large datasets given as matrix of samples times features and has been successfully applied in life sciences and e-commerce for drug design and recommender systems, respectively. actor nalysis for cluster cquisition (FABIA), one of the most successful biclustering methods, is a generative model that represents each bicluster by two sparse membership vectors: one for the samples and one for the features. However, FABIA is restricted to about 20 code units because of the high computational complexity of computing the posterior. Furthermore, code units are sometimes insufficiently decorrelated and sample membership is difficult to determine. We propose to use the recently introduced unsupervised Deep Learning approach Rectified Factor Networks (RFNs) to overcome the drawbacks of existing biclustering methods. RFNs efficiently construct very sparse, non-linear, high-dimensional representations of the input via their posterior means. RFN learning is a generalized alternating minimization algorithm based on the posterior regularization method which enforces non-negative and normalized posterior means. Each code unit represents a bicluster, where samples for which the code unit is active belong to the bicluster and features that have activating weights to the code unit belong to the bicluster.

RESULTS

On 400 benchmark datasets and on three gene expression datasets with known clusters, RFN outperformed 13 other biclustering methods including FABIA. On data of the 1000 Genomes Project, RFN could identify DNA segments which indicate, that interbreeding with other hominins starting already before ancestors of modern humans left Africa.

AVAILABILITY AND IMPLEMENTATION

https://github.com/bioinf-jku/librfn.

CONTACT

djork-arne.clevert@bayer.com or hochreit@bioinf.jku.at.

Identification of Non-Electrophilic Nrf2 Activators from Approved Drugs

Oxidative damage can lead to a wide range of diseases. Nrf2 is an important transcription factor that regulates many of the cytoprotective enzymes involved in the oxidative stress response. Therefore, targeting the regulation of Nrf2 activation is one logical and effective strategy to prevent or lower the risk of oxidative stress-related diseases. Until now, most research has focused on electrophilic indirect Nrf2 activators, but the risk of 'off-target' effects may be associated with these activators. To find novel small non-electrophilic modulators of Nrf2, we started from chemical agents derived from a connectivity map (cMap) and identified 22 non-electrophilic potential Nrf2-activating drugs through a drug repositioning tactic. By determining the expression changes of antioxidant genes in MCF7 cells that were treated with the potential Nrf2 activators using quantitative real-time polymerase chain reaction RT-PCR (real-time polymerase chain reaction) (qRT-PCR), astemizole was found to have a greater scale of upregulating antioxidant genes NQO1, HO-1, and GCLM than the positive control d,l-sulforaphane, although the testing concentration was lower than that of the control. Astemizole is a good potential redox regulator and deserves more pharmacodynamic experimentation to test and verify its feasibility for use as an Nrf2 activator.

Quantitative Identification of Compound-Dependent On-Modules and Differential Allosteric Modules From Homologous Ischemic Networks

Module‐based methods have made much progress in deconstructing biological networks. However, it is a great challenge to quantitatively compare the topological structural variations of modules (allosteric modules [AMs]) under different situations. A total of 23, 42, and 15 coexpression modules were identified in baicalin (BA), jasminoidin (JA), and ursodeoxycholic acid (UA) in a global anti‐ischemic mice network, respectively. Then, we integrated the methods of module‐based consensus ratio (MCR) and modified Z_summary module statistic to validate 12 BA, 22 JA, and 8 UA on‐modules based on comparing with vehicle. The MCRs for pairwise comparisons were 1.55% (BA vs. JA), 1.45% (BA vs. UA), and 1.27% (JA vs. UA), respectively. Five conserved allosteric modules (CAMs) and 17 unique allosteric modules (UAMs) were identified among these groups. In conclusion, module‐centric analysis may provide us a unique approach to understand multiple pharmacological mechanisms associated with differential phenotypes in the era of modular pharmacology.

Why Polyphenols have Promiscuous Actions? An Investigation by Chemical Bioinformatics

Despite their diverse pharmacological effects, polyphenols are poor for use as drugs, which have been traditionally ascribed to their low bioavailability. However, Baell and co-workers recently proposed that the redox potential of polyphenols also plays an important role in this, because redox reactions bring promiscuous actions on various protein targets and thus produce non-specific pharmacological effects. To investigate whether the redox reactivity behaves as a critical factor in polyphenol promiscuity, we performed a chemical bioinformatics analysis on the structure-activity relationships of twenty polyphenols. It was found that the gene expression profiles of human cell lines induced by polyphenols were not correlated with the presence or not of redox moieties in the polyphenols, but significantly correlated with their molecular structures. Therefore, it is concluded that the promiscuous actions of polyphenols are likely to result from their inherent structural features rather than their redox potential.

Network pharmacology applications to map the unexplored target space and therapeutic potential of natural products

It is widely accepted that drug discovery often requires a systems-level polypharmacology approach to tackle problems such as lack of efficacy and emerging resistance of single-targeted compounds. Network pharmacology approaches are increasingly being developed and applied to find new therapeutic opportunities and to re-purpose approved drugs. However, these recent advances have been relatively slow to be translated into the field of natural products. Here, we argue that a network pharmacology approach would enable an effective mapping of the yet unexplored target space of natural products, hence providing a systematic means to extend the druggable space of proteins implicated in various complex diseases. We give an overview of the key network pharmacology concepts and recent experimental-computational approaches that have been successfully applied to natural product research, including unbiased elucidation of mechanisms of action as well as systematic prediction of effective therapeutic combinations. We focus specifically on anticancer applications that use in vivo and in vitro functional phenotypic measurements, such as genome-wide transcriptomic response profiles, which enable a global modelling of the multi-target activity at the level of the biological pathways and interaction networks. We also provide representative examples of other disease applications, databases and tools as well as existing and emerging resources, which may prove useful for future natural product research. Finally, we offer our personal view of the current limitations, prospective developments and open questions in this exciting field.

Elucidating pharmacological mechanisms of natural medicines by biclustering analysis of the gene expression profile: a case study on curcumin and Si-Wu-Tang

Natural medicines have attracted wide attention in recent years. It is of great significance to clarify the pharmacological mechanisms of natural medicines. In prior studies, we established a method for elucidating pharmacological mechanisms of natural products contained in connectivity map (cMap), in terms of module profiles of gene expression in chemical treatments. In this study, we explore whether this methodology is applicable to dissecting the pharmacological mechanisms of natural medicines beyond the agents contained in cMap. First, the gene expression profiles of curcumin (a typical isolated natural medicine) and Si-Wu-Tang (a classic traditional Chinese medicine formula) treatments were merged with those of cMap-derived 1309 agents, respectively. Then, a biclustering analysis was performed using FABIA method to identify gene modules. The biological functions of gene modules provide preliminary insights into pharmacological mechanisms of both natural medicines. The module profile can be characterized by a binary vector, which allowed us to compare the expression profiles of natural medicines with those of cMap-derived agents. Accordingly, we predicted a series of pharmacological effects for curcumin and Si-Wu-Tang by the indications of cMap-covered drugs. Most predictions were supported by experimental observations, suggesting the potential use of this method in natural medicine dissection.

Elucidating polypharmacological mechanisms of polyphenols by gene module profile analysis

Due to the diverse medicinal effects, polyphenols are among the most intensively studied natural products. However, it is a great challenge to elucidate the polypharmacological mechanisms of polyphenols. To address this challenge, we establish a method for identifying multiple targets of chemical agents through analyzing the module profiles of gene expression upon chemical treatments. By using FABIA algorithm, we have performed a biclustering analysis of gene expression profiles derived from Connectivity Map (cMap), and clustered the profiles into 49 gene modules. This allowed us to define a 49 dimensional binary vector to characterize the gene module profiles, by which we can compare the expression profiles for each pair of chemical agents with Tanimoto coefficient. For the agent pairs with similar gene expression profiles, we can predict the target of one agent from the other. Drug target enrichment analysis indicated that this method is efficient to predict the multiple targets of chemical agents. By using this method, we identify 148 targets for 20 polyphenols derived from cMap. A large part of the targets are validated by experimental observations. The results show that the medicinal effects of polyphenols are far beyond their well-known antioxidant activities. This method is also applicable to dissect the polypharmacology of other natural products.