A survey of current trends in computational drug repositioning

Differential associations of depression-related phenotypes with cardiometabolic risks: Polygenic analyses and exploring shared genetic variants and pathways

BACKGROUND

Numerous studies have suggested associations between depression and cardiometabolic (CM) diseases. However, little is known about the mechanism underlying this comorbidity, and whether the relationship differs by depression subtypes.

METHODS

Using polygenic risk scores (PRS) and linkage disequilibrium (LD) score regression, we investigated the genetic overlap of various depression-related phenotypes with a comprehensive panel of 20 CM traits. GWAS results for major depressive disorder (MDD) were taken from the PGC and CONVERGE studies, with the latter focusing on severe melancholic depression. GWAS results on general depressive symptoms (DS) and neuroticism were also included. We identified the shared genetic variants and inferred enriched pathways. We also looked for drugs over-represented among the top-shared genes, with an aim to finding repositioning opportunities for comorbidities.

RESULTS

We found significant genetic overlap between MDD, DS, and neuroticism with cardiometabolic traits. In general, positive polygenic associations with CM abnormalities were observed except for MDD-CONVERGE. Counterintuitively, PRS representing severe melancholic depression was associated with reduced CM risks. Enrichment analyses of shared SNPs revealed many interesting pathways such as those related to inflammation that underlie the comorbidity of depressive and CM traits. Using a gene-set analysis approach, we also revealed several repositioning candidates with literature support (e.g., bupropion).

CONCLUSIONS

Our study highlights shared genetic bases of depression with CM traits, and suggests the associations vary by depression subtypes, which may have implications in targeted prevention of cardiovascular events for patients. Identification of shared genetic factors may also guide drug discovery for the comorbidities.

Progress Toward Cancer Data Ecosystems

One of the recommendations of the Cancer Moonshot Blue Ribbon Panel report from 2016 was the creation of a national cancer data ecosystem. We review some of the approaches for building cancer data ecosystems and some of the progress that has been made. A data commons is the colocation of data with cloud computing infrastructure and commonly used software services, tools, and applications for managing, integrating, analyzing, and sharing data to create an interoperable resource for the research community. We discuss data commons and their potential role in cancer data ecosystems and, in particular, how multiple data commons can interoperate to form part of the foundation for a cancer data ecosystem.

Drug repositioning through integration of prior knowledge and projections of drugs and diseases

Motivation

Identifying and developing novel therapeutic effects for existing drugs contributes to reduction of drug development costs. Most of the previous methods focus on integration of the heterogeneous data of drugs and diseases from multiple sources for predicting the candidate drug–disease associations. However, they fail to take the prior knowledge of drugs and diseases and their sparse characteristic into account. It is essential to develop a method that exploits the more useful information to predict the reliable candidate associations.

Results

We present a method based on non-negative matrix factorization, DisDrugPred, to predict the drug-related candidate disease indications. A new type of drug similarity is firstly calculated based on their associated diseases. DisDrugPred completely integrates two types of disease similarities, the associations between drugs and diseases, and the various similarities between drugs from different levels including the chemical structures of drugs, the target proteins of drugs, the diseases associated with drugs and the side effects of drugs. The prior knowledge of drugs and diseases and the sparse characteristic of drug–disease associations provide a deep biological perspective for capturing the relationships between drugs and diseases. Simultaneously, the possibility that a drug is associated with a disease is also dependant on their projections in the low-dimension feature space. Therefore, DisDrugPred deeply integrates the diverse prior knowledge, the sparse characteristic of associations and the projections of drugs and diseases. DisDrugPred achieves superior prediction performance than several state-of-the-art methods for drug–disease association prediction. During the validation process, DisDrugPred also can retrieve more actual drug–disease associations in the top part of prediction result which often attracts more attention from the biologists. Moreover, case studies on five drugs further confirm DisDrugPred’s ability to discover potential candidate disease indications for drugs.

Availability and implementation

The fourth type of drug similarity and the predicted candidates for all the drugs are available at https://github.com/pingxuan-hlju/DisDrugPred.

Supplementary information

Supplementary data are available at Bioinformatics online.

A systems-level analysis of drug-target-disease associations for drug repositioning

Drug repositioning is the process of finding new therapeutic uses for existing, approved drugs-a process thathas value when considering the exorbitant costs of novel drug development. Several computational strategies exist as a way to predict these alternative applications. In this study, we used datasets on: (1) human biological drug targets and (2) disease-associated genes and, based on a direct functional interaction between them, searched for potential opportunities for drug repositioning. From the set of 1125 unique drug targets and their 88 490 interactions with disease-associated genes, 30 drug targets were analyzed and (3) discussed in detail for the purpose of this article. The current indications of the drugs thattarget them were validated through the interactions, and new opportunities for repositioning were predicted. Among the set of drugs for potential repositioning werebenzodiazepines for the treatment of autism spectrum disorders; nortriptyline for the treatment of melanoma, glioma and other cancers; and vitamin B6 in prevention of spontaneous abortions and cleft palate birth defects. Special emphasis was also placed on those new potential indications that pertained to orphan diseases-these are diseases whose rarity means that development of novel treatment is not financially viable. This computational drug repositioning approach uses existing information on drugs and drug targets, and insights into the genetic basis of disease, as a means to systematically generate the most probable new uses for the drugs on offer, and in this way harness their true therapeutic power.

Re-positioning of known drugs for Pim-1 kinase target using molecular docking analysis

The Concept of reusing existing drugs for new targets is gaining momentum in recent years because of cost-effectiveness as safety and toxicology data are already available. Therefore, it is of interest to re-profile known drugs against the Pim-1 kinase target using molecular docking analysis. Results show that known drugs such as nilotinib, vemurafenib, Idelalisib, and other small kinases inhibitors have high binding ability with Pim-1 kinase for consideration as potential inhibitors.

Synergy from gene expression and network mining (SynGeNet) method predicts synergistic drug combinations for diverse melanoma genomic subtypes

Systems biology perspectives are crucial for understanding the pathophysiology of complex diseases, and therefore hold great promise for the discovery of novel treatment strategies. Drug combinations have been shown to improve durability and reduce resistance to available first-line therapies in a variety of cancers; however, traditional drug discovery approaches are prohibitively cost and labor-intensive to evaluate large-scale matrices of potential drug combinations. Computational methods are needed to efficiently model complex interactions of drug target pathways and identify mechanisms underlying drug combination synergy. In this study, we employ a computational approach, SynGeNet (Synergy from Gene expression and Network mining), which integrates transcriptomics-based connectivity mapping and network centrality analysis to analyze disease networks and predict drug combinations. As an exemplar of a disease in which combination therapies demonstrate efficacy in genomic-specific contexts, we investigate malignant melanoma. We employed SynGeNet to generate drug combination predictions for each of the four major genomic subtypes of melanoma (BRAF, NRAS, NF1, and triple wild type) using publicly available gene expression and mutation data. We validated synergistic drug combinations predicted by our method across all genomic subtypes using results from a high-throughput drug screening study across. Finally, we prospectively validated the drug combination for BRAF-mutant melanoma that was top ranked by our approach, vemurafenib (BRAF inhibitor) + tretinoin (retinoic acid receptor agonist), using both in vitro and in vivo models of BRAF-mutant melanoma and RNA-sequencing analysis of drug-treated melanoma cells to validate the predicted mechanisms. Our approach is applicable to a wide range of disease domains, and, importantly, can model disease-relevant protein subnetworks in precision medicine contexts.

Computational Drug Repositioning for Gastric Cancer using Reversal Gene Expression Profiles

Treatment of gastric cancer (GC) often produces poor outcomes. Moreover, predicting which GC treatments will be effective remains challenging. Computational drug repositioning using public databases is a promising and efficient tool for discovering new uses for existing drugs. Here we used a computational reversal of gene expression approach based on effects on gene expression signatures by GC disease and drugs to explore new GC drug candidates. Gene expression profiles for individual GC tumoral and normal gastric tissue samples were downloaded from the Gene Expression Omnibus (GEO) and differentially expressed genes (DEGs) in GC were determined with a meta-signature analysis. Profiles drug activity and drug-induced gene expression were downloaded from the ChEMBL and the LINCS databases, respectively. Candidate drugs to treat GC were predicted using reversal gene expression score (RGES). Drug candidates including sorafenib, olaparib, elesclomol, tanespimycin, selumetinib, and ponatinib were predicted to be active for treatment of GC. Meanwhile, GC-related genes such as PLOD3, COL4A1, UBE2C, MIF, and PRPF5 were identified as having gene expression profiles that can be reversed by drugs. These findings support the use of a computational reversal gene expression approach to identify new drug candidates that can be used to treat GC.

In silico drug repositioning based on drug-miRNA associations

Drug repositioning has become a prevailing tactic as this strategy is efficient, economical and low risk for drug discovery. Meanwhile, recent studies have confirmed that small-molecule drugs can modulate the expression of disease-related miRNAs, which indicates that miRNAs are promising therapeutic targets for complex diseases. In this study, we put forward and verified the hypothesis that drugs with similar miRNA profiles may share similar therapeutic properties. Furthermore, a comprehensive drug–drug interaction network was constructed based on curated drug-miRNA associations. Through random network comparison, topological structure analysis and network module extraction, we found that the closely linked drugs in the network tend to treat the same diseases. Additionally, the curated drug–disease relationships (from the CTD) and random walk with restarts algorithm were utilized on the drug–drug interaction network to identify the potential drugs for a given disease. Both internal validation (leave-one-out cross-validation) and external validation (independent drug–disease data set from the ChEMBL) demonstrated the effectiveness of the proposed approach. Finally, by integrating drug-miRNA and miRNA-disease information, we also explain the modes of action of drugs in the view of miRNA regulation. In summary, our work could determine novel and credible drug indications and offer novel insights and valuable perspectives for drug repositioning.

Synergistic drug combinations prediction by integrating pharmacological data

There is compelling evidence that synergistic drug combinations have become promising strategies for combating complex diseases, and they have evident predominance comparing to traditional one drug - one disease approaches. In this paper, we develop a computational method, namely SyFFM, that takes pharmacological data into consideration and applies field-aware factorization machines to analyze and predict potential synergistic drug combinations. Firstly, features of drug pairs are constructed based on associations between drugs and target, and enzymes, and indication areas. Then, the synergistic scores of drug combinations are obtained by implementing field-aware factorization machines on latent vector space of these features. Finally, synergistic combinations can be predicted by introducing a threshold. We applied SyFFM to predict pairwise synergistic combinations and three-drug synergistic combinations, and the performance is good in terms of cross-validation. Besides, more than 90% combinations of the top ranked predictions are proved by literature and the analysis of parameters in model shows that our method can help to investigate and explain synergistic mechanisms underlying combinatorial therapy.

Methods for Discovering and Targeting Druggable Protein-Protein Interfaces and Their Application to Repurposing

Drug repurposing is a creative and resourceful approach to increase the number of therapies by exploiting available and approved drugs. However, identifying new protein targets for previously approved drugs is challenging. Although new strategies have been developed for drug repurposing, there is broad agreement that there is room for further improvements. In this chapter, we review protein-protein interaction (PPI) interface-targeting strategies for drug repurposing applications. We discuss certain features, such as hot spot residue and hot region prediction and their importance in drug repurposing, and illustrate common methods used in PPI networks to identify drug off-targets. We also collect available online resources for hot spot prediction, binding pocket identification, and interface clustering which are effective resources in polypharmacology. Finally, we provide case studies showing the significance of protein interfaces and hot spots in drug repurposing.

Repurposing Drugs Based on Evolutionary Relationships Between Targets of Approved Drugs and Proteins of Interest

Drug repurposing has garnered much interest as an effective method for drug development among biopharmaceutical companies. The availability of information on complete sequences of genomes and their associated biological data, genotype-phenotype-disease relationships, and properties of small molecules offers opportunities to explore the repurpose-able potential of existing pharmacopoeia. This method gains further importance, especially, in the context of development of drugs against infectious diseases, some of which pose serious complications due to emergence of drug-resistant pathogens. In this article, we describe computational means to achieve potential repurpose-able drug candidates that may be used against infectious diseases by exploring evolutionary relationships between established targets of FDA-approved drugs and proteins of pathogen of interest.

Precision drug repurposing via convergent eQTL-based molecules and pathway targeting independent disease-associated polymorphisms

Repurposing existing drugs for new therapeutic indications can improve success rates and streamline development. Use of large-scale biomedical data repositories, including eQTL regulatory relationships and genome-wide disease risk associations, offers opportunities to propose novel indications for drugs targeting common or convergent molecular candidates associated to two or more diseases. This proposed novel computational approach scales across 262 complex diseases, building a multi-partite hierarchical network integrating (i) GWAS-derived SNP-to-disease associations, (ii) eQTL-derived SNP-to-eGene associations incorporating both cis- and trans-relationships from 19 tissues, (iii) protein target-to-drug, and (iv) drug-to-disease indications with (iv) Gene Ontology-based information theoretic semantic (ITS) similarity calculated between protein target functions. Our hypothesis is that if two diseases are associated to a common or functionally similar eGene - and a drug targeting that eGene/protein in one disease exists - the second disease becomes a potential repurposing indication. To explore this, all possible pairs of independently segregating GWAS-derived SNPs were generated, and a statistical network of similarity within each SNP-SNP pair was calculated according to scale-free overrepresentation of convergent biological processes activity in regulated eGenes (ITSeGENE-eGENE) and scale-free overrepresentation of common eGene targets between the two SNPs (ITSSNP-SNP). Significance of ITSSNP-SNP was conservatively estimated using empirical scale-free permutation resampling keeping the node-degree constant for each molecule in each permutation. We identified 26 new drug repurposing indication candidates spanning 89 GWAS diseases, including a potential repurposing of the calcium-channel blocker Verapamil from coronary disease to gout. Predictions from our approach are compared to known drug indications using DrugBank as a gold standard (odds ratio=13.1, p-value=2.49x10-8). Because of specific disease-SNPs associations to candidate drug targets, the proposed method provides evidence for future precision drug repositioning to a patient's specific polymorphisms.

A Drug Repurposing Method Based on Drug-Drug Interaction Networks and Using Energy Model Layouts

Complex network representations of reported drug-drug interactions foster computational strategies that can infer pharmacological functions which, in turn, create incentives for drug repositioning. Here, we use Gephi (a platform for complex network visualization and analysis) to represent a drug-drug interaction network with drug interaction information from DrugBank 4.1. Both modularity class- and force-directed layout ForceAtlas2 are employed to generate drug clusters which correspond to nine specific drug properties. Most drugs comply with their cluster's dominant property; however, some of them seem not to be in a proper position (i.e., in accordance with their already known functions). Such cases, along with cases of drugs that are topologically placed in the overlapping or bordering zones between clusters, may indicate previously unaccounted pharmacologic functions, thus leading to potential repositionings. Out of the 1141 drugs with relevant information on their interactions in DrugBank 4.1, we confirm the predicted properties for 85% of the drugs. The high prediction rate of our methodology suggests that, at least for some of the 15% drugs that seem to be inconsistent with the predicted property, we can get very good repositioning hints. As such, we present illustrative examples of recovered well-known repositionings, as well as recently confirmed pharmacological properties.

Computational drug repositioning using meta-path-based semantic network analysis

BACKGROUND

Drug repositioning is a promising and efficient way to discover new indications for existing drugs, which holds the great potential for precision medicine in the post-genomic era. Many network-based approaches have been proposed for drug repositioning based on similarity networks, which integrate multiple sources of drugs and diseases. However, these methods may simply view nodes as the same-typed and neglect the semantic meanings of different meta-paths in the heterogeneous network. Therefore, it is urgent to develop a rational method to infer new indications for approved drugs.

RESULTS

In this study, we proposed a novel methodology named HeteSim_DrugDisease (HSDD) for the prediction of drug repositioning. Firstly, we build the drug-drug similarity network and disease-disease similarity network by integrating the information of drugs and diseases. Secondly, a drug-disease heterogeneous network is constructed, which combines the drug similarity network, disease similarity network as well as the known drug-disease association network. Finally, HSDD predicts novel drug-disease associations based on the HeteSim scores of different meta-paths. The experimental results show that HSDD performs significantly better than the existing state-of-the-art approaches. HSDD achieves an AUC score of 0.8994 in the leave-one-out cross validation experiment. Moreover, case studies for selected drugs further illustrate the practical usefulness of HSDD.

CONCLUSIONS

HSDD can be an effective and feasible way to infer the associations between drugs and diseases using on meta-path-based semantic network analysis.

DDR: efficient computational method to predict drug-target interactions using graph mining and machine learning approaches

Motivation

Finding computationally drug–target interactions (DTIs) is a convenient strategy to identify new DTIs at low cost with reasonable accuracy. However, the current DTI prediction methods suffer the high false positive prediction rate.

Results

We developed DDR, a novel method that improves the DTI prediction accuracy. DDR is based on the use of a heterogeneous graph that contains known DTIs with multiple similarities between drugs and multiple similarities between target proteins. DDR applies non-linear similarity fusion method to combine different similarities. Before fusion, DDR performs a pre-processing step where a subset of similarities is selected in a heuristic process to obtain an optimized combination of similarities. Then, DDR applies a random forest model using different graph-based features extracted from the DTI heterogeneous graph. Using 5-repeats of 10-fold cross-validation, three testing setups, and the weighted average of area under the precision-recall curve (AUPR) scores, we show that DDR significantly reduces the AUPR score error relative to the next best start-of-the-art method for predicting DTIs by 31% when the drugs are new, by 23% when targets are new and by 34% when the drugs and the targets are known but not all DTIs between them are not known. Using independent sources of evidence, we verify as correct 22 out of the top 25 DDR novel predictions. This suggests that DDR can be used as an efficient method to identify correct DTIs.

Availability and implementation

The data and code are provided at https://bitbucket.org/RSO24/ddr/.

Supplementary information

Supplementary data are available at Bioinformatics online.

Network embedding in biomedical data science

Owning to the rapid development of computer technologies, an increasing number of relational data have been emerging in modern biomedical research. Many network-based learning methods have been proposed to perform analysis on such data, which provide people a deep understanding of topology and knowledge behind the biomedical networks and benefit a lot of applications for human healthcare. However, most network-based methods suffer from high computational and space cost. There remain challenges on handling high dimensionality and sparsity of the biomedical networks. The latest advances in network embedding technologies provide new effective paradigms to solve the network analysis problem. It converts network into a low-dimensional space while maximally preserves structural properties. In this way, downstream tasks such as link prediction and node classification can be done by traditional machine learning methods. In this survey, we conduct a comprehensive review of the literature on applying network embedding to advance the biomedical domain. We first briefly introduce the widely used network embedding models. After that, we carefully discuss how the network embedding approaches were performed on biomedical networks as well as how they accelerated the downstream tasks in biomedical science. Finally, we discuss challenges the existing network embedding applications in biomedical domains are faced with and suggest several promising future directions for a better improvement in human healthcare.

Disease comorbidity-guided drug repositioning: a case study in schizophrenia

The key to any computational drug repositioning is the availability of relevant data in machine-understandable format. While large amount of genetic, genomic and chemical data are publicly available, large-scale higher-level disease and drug phenotypic data are limited. We recently constructed a large-scale disease-comorbidity relationship knowledge base (dCombKB) and a comprehensive drug-treatment relationship knowledge base (TreatKB) from 21 million biomedical research articles and other resources. In this study, we demonstrated the potential of dCombKB and TreatKB in drug repositioning for schizophrenia, one of the top ten illnesses contributing to the global burden of disease. dCombKB contains 121,359 unique disease-disease comorbidity pairs for 23,041 diseases. TreatKB contains 208,330 unique drug-disease treatment pairs for 2,484 drugs and 24,511 diseases. We constructed a phenotypic comorbidity disease network (PDN) of 14,645 disease nodes and 101,275 edges based on dCombKB. We applied standard network-based ranking algorithm to find diseases that are phenotypically related to SCZ. We developed a drug prioritization system, PhenoPredict-CDN, to systematically reposition drugs for SCZ from diseases phenotypically related to SCZ. PhenoPredict-CDN found all 18 FDA-approved SCZ drugs and ranked them highly as tested in a de-novo validation setting (recall: 1.0, mean ranking: top 6.05%, median ranking: top 1.65%). When compared to PREDICT, a comprehensive drug repositioning system, for novel predictions, Pheno-Predict-CDN outperformed PREDICT in Precision-Recall (PR) curves across three different evaluation datasets. Compared to PREDICT, PhenoPredict-CDN showed a significant 110.0-230.0% improvements in mean average precision. In summary, large-scale higher-level disease-comorbidity relationships data extracted from biomedical literature has potential in drug discovery for SCZ, a complex disease with unknown pathophysiological mechanisms. All the data are publicly available: dCombKB at http://nlp.

CASE

edu/public/data/dCombKB, TreatKB at http://nlp.

CASE

edu/public/data/treatKB/, and predictions for SCZ at http://nlp.

CASE

edu/public/data/SCZ_CDN/.

Retrospective analysis of health claims to evaluate pharmacotherapies with potential for repurposing: Association of bupropion and stimulant use disorder remission

Drug repurposing is the identification of novel indication(s) for existing medications. Health claims data provide a burgeoning resource to evaluate pharmacotherapies with repurposing potential. To demonstrate a workflow for drug repurposing using claims data, we assessed the association between prescription of bupropion and stimulant use disorder (StUD) remission. Using the Truven Marketscan database, 96,156 individuals with a StUD were identified. Logistic regression was used to model the association between new bupropion prescriptions and remission while controlling for age, sex, region, StUD severity, antidepressant co-prescriptions, and comorbid mood and attention disorders. Prescription of bupropion within 30 days offirst documented StUD diagnosis increased odds of a subsequent remission diagnosis by 2.1 times (99% confidence interval: 1.09-3.89) in individuals with an amphetamine use disorder, but not those with a cocaine use disorder. This work provides a framework for reverse-translational drug repurposing, which may be applied to many other medical conditions.

Adverse Drug Reaction Predictions Using Stacking Deep Heterogeneous Information Network Embedding Approach

Inferring potential adverse drug reactions is an important and challenging task for the drug discovery and healthcare industry. Many previous studies in computational pharmacology have proposed utilizing multi-source drug information to predict drug side effects have and achieved initial success. However, most of the prediction methods mainly rely on direct similarities inferred from drug information and cannot fully utilize the drug information about the impact of protein–protein interactions (PPI) on potential drug targets. Moreover, most of the methods are designed for specific tasks. In this work, we propose a novel heterogeneous network embedding approach for learning drug representations called SDHINE, which integrates PPI information into drug embeddings and is generic for different adverse drug reaction (ADR) prediction tasks. To integrate heterogeneous drug information and learn drug representations, we first design different meta-path-based proximities to calculate drug similarities, especially target propagation meta-path-based proximity based on PPI network, and then construct a semi-supervised stacking deep neural network model that is jointly optimized by the defined meta-path proximities. Extensive experiments with three state-of-the-art network embedding methods on three ADR prediction tasks demonstrate the effectiveness of the SDHINE model. Furthermore, we compare the drug representations in terms of drug differentiation by mapping the representations into 2D space; the results show that the performance of our approach is superior to that of the comparison methods.

Drug repositioning using drug-disease vectors based on an integrated network

BACKGROUND

Diverse interactions occur between biomolecules, such as activation, inhibition, expression, or repression. However, previous network-based studies of drug repositioning have employed interaction on the binary protein-protein interaction (PPI) network without considering the characteristics of the interactions. Recently, some studies of drug repositioning using gene expression data found that associations between drug and disease genes are useful information for identifying novel drugs to treat diseases. However, the gene expression profiles for drugs and diseases are not always available. Although gene expression profiles of drugs and diseases are available, existing methods cannot use the drugs or diseases, when differentially expressed genes in the profiles are not included in their network.

RESULTS

We developed a novel method for identifying candidate indications of existing drugs considering types of interactions between biomolecules based on known drug-disease associations. To obtain associations between drug and disease genes, we constructed a directed network using protein interaction and gene regulation data obtained from various public databases providing diverse biological pathways. The network includes three types of edges depending on relationships between biomolecules. To quantify the association between a target gene and a disease gene, we explored the shortest paths from the target gene to the disease gene and calculated the types and weights of the shortest paths. For each drug-disease pair, we built a vector consisting of values for each disease gene influenced by the drug. Using the vectors and known drug-disease associations, we constructed classifiers to identify novel drugs for each disease.

CONCLUSION

We propose a method for exploring candidate drugs of diseases using associations between drugs and disease genes derived from a directed gene network instead of gene regulation data obtained from gene expression profiles. Compared to existing methods that require information on gene relationships and gene expression data, our method can be applied to a greater number of drugs and diseases. Furthermore, to validate our predictions, we compared the predictions with drug-disease pairs in clinical trials using the hypergeometric test, which showed significant results. Our method also showed better performance compared to existing methods for the area under the receiver operating characteristic curve (AUC).

A Review of Drug Repositioning Based Chemical-induced Cell Line Expression Data

Drug repositioning is an important area of biomedical research. The drug repositioning studies have shifted to computational approaches. Large-scale perturbation databases, such as the Connectivity Map and the Library of Integrated Network-Based Cellular Signatures, contain a number of chemical-induced gene expression profiles and provide great opportunities for computational biology and drug repositioning. One reason is that the profiles provided by the Connectivity Map and the Library of Integrated Network-Based Cellular Signatures databases show an overall view of biological mechanism in drugs, diseases and genes. In this article, we provide a review of the two databases and their recent applications in drug repositioning.

A review of validation strategies for computational drug repositioning

Repositioning of previously approved drugs is a promising methodology because it reduces the cost and duration of the drug development pipeline and reduces the likelihood of unforeseen adverse events. Computational repositioning is especially appealing because of the ability to rapidly screen candidates in silico and to reduce the number of possible repositioning candidates. What is unclear, however, is how useful such methods are in producing clinically efficacious repositioning hypotheses. Furthermore, there is no agreement in the field over the proper way to perform validation of in silico predictions, and in fact no systematic review of repositioning validation methodologies. To address this unmet need, we review the computational repositioning literature and capture studies in which authors claimed to have validated their work. Our analysis reveals widespread variation in the types of strategies, predictions made and databases used as ‘gold standards’. We highlight a key weakness of the most commonly used strategy and propose a path forward for the consistent analytic validation of repositioning techniques.

Quantitative and systems pharmacology 4. Network-based analysis of drug pleiotropy on coronary artery disease

Despite recent advance of therapeutic development, coronary artery disease (CAD) remains one of the major issues to public health. The use of genomics and systems biology approaches to inform drug discovery and development have offered the possibilities for new target identification and in silico drug repurposing. In this study, we propose a network-based, systems pharmacology framework for target identification and drug repurposing in pharmacologic treatment and chemoprevention of CAD. Specifically, we build in silico models by integrating known drug-target interactions, CAD genes derived from the genetic and genomic studies, and the human protein-protein interactome. We demonstrate that the proposed in silico models can successfully uncover approved drugs and novel natural products in potentially treating and preventing CAD. In case studies, we highlight several approved drugs (e.g., fasudil, parecoxib, and dexamethasone) or natural products (e.g., resveratrol, luteolin, daidzein and caffeic acid) with new mechanism-of-action in chemical intervention of CAD by network analysis. In summary, this study offers a powerful systems pharmacology approach for target identification and in silico drug repurposing on CAD.

An integrative approach using real-world data to identify alternative therapeutic uses of existing drugs

Different computational approaches are employed to efficiently identify novel repositioning possibilities utilizing different sources of information and algorithms. It is critical to propose high-valued candidate-repositioning possibilities before conducting lengthy in vivo validation studies that consume significant resources. Here we report a novel multi-methodological approach to identify opportunities for drug repositioning. We performed analyses of real-world data (RWD) acquired from the United States Food and Drug Administration’s Adverse Event Reporting System (FAERS) and the claims database maintained by the Japan Medical Data Center (JMDC). These analyses were followed by cross-validation through bioinformatics analyses of gene expression data. Inverse associations revealed using disproportionality analysis (DPA) and sequence symmetry analysis (SSA) were used to detect potential drug-repositioning signals. To evaluate the validity of the approach, we conducted a feasibility study to identify marketed drugs with the potential for treating inflammatory bowel disease (IBD). Primary analyses of the FAERS and JMDC claims databases identified psycholeptics such as haloperidol, diazepam, and hydroxyzine as candidates that may improve the treatment of IBD. To further investigate the mechanistic relevance between hit compounds and disease pathology, we conducted bioinformatics analyses of the associations of the gene expression profiles of these compounds with disease. We identified common biological features among genes differentially expressed with or without compound treatment as well as disease-perturbation data available from open sources, which strengthened the mechanistic rationale of our initial findings. We further identified pathways such as cytokine signaling that are influenced by these drugs. These pathways are relevant to pathologies and can serve as alternative targets of therapy. Integrative analysis of RWD such as those available from adverse-event databases, claims databases, and transcriptome analyses represent an effective approach that adds value to efficiently identifying potential novel therapeutic opportunities.

PISTON: Predicting drug indications and side effects using topic modeling and natural language processing

The process of discovering novel drugs to treat diseases requires a long time and high cost. It is important to understand side effects of drugs as well as their therapeutic effects, because these can seriously damage the patients due to unexpected actions of the derived candidate drugs. In order to overcome these limitations, computational methods for predicting the therapeutic effects and side effects have been proposed. In particular, text mining is a widely used technique in the field of systems biology, because it can discover hidden relationships between drugs, genes and diseases from a large amount of literature data. Compared with in vivo/in vitro experiments, text mining derives meaningful results with less time and cost. In this study, we propose an algorithm for predicting novel drug-phenotype associations and drug-side effect associations using topic modeling and natural language processing (NLP). We extract sentences in which drugs and genes co-occur from the abstracts of the literature and identify words that describe the relationship between them using NLP. Considering the characteristics of the identified words, we determine if the drug has an up-regulation effect or a down-regulation effect on the gene. Based on genes that affect drugs and their regulatory relationships, we group the frequently occurring genes and regulatory relationships into topics, and build a drug-topic probability matrix by calculating the score that the drug will have a topic using topic modeling. Using the matrix, a classifier is constructed for predicting the novel indications and side effects of drugs considering the characteristics of known drug-phenotype associations or drug-side effect associations. The proposed method predicts both indications and side effects with a single algorithm, and it can exclude drugs with serious side effects or side effects that patients do not want to experience from among the candidate drugs provided for the treatment of the phenotype. Furthermore, lists of novel candidate drugs for phenotypes and side effects can be continuously updated with our algorithm every time a document is added. More than a thousand documents are produced per day, and it is possible for our algorithm to efficiently derive candidate drugs because it requires less cost than the existing drug repositioning methods. The resource of PISTON is available at databio.gachon.ac.kr/tools/PISTON.

Machine Learning for Drug-Target Interaction Prediction

Identifying drug-target interactions will greatly narrow down the scope of search of candidate medications, and thus can serve as the vital first step in drug discovery. Considering that in vitro experiments are extremely costly and time-consuming, high efficiency computational prediction methods could serve as promising strategies for drug-target interaction (DTI) prediction. In this review, our goal is to focus on machine learning approaches and provide a comprehensive overview. First, we summarize a brief list of databases frequently used in drug discovery. Next, we adopt a hierarchical classification scheme and introduce several representative methods of each category, especially the recent state-of-the-art methods. In addition, we compare the advantages and limitations of methods in each category. Lastly, we discuss the remaining challenges and future outlook of machine learning in DTI prediction. This article may provide a reference and tutorial insights on machine learning-based DTI prediction for future researchers.

Review and comparative assessment of similarity-based methods for prediction of drug–protein interactions in the druggable human proteome

Drug–protein interactions (DPIs) underlie the desired therapeutic actions and the adverse side effects of a significant majority of drugs. Computational prediction of DPIs facilitates research in drug discovery, characterization and repurposing. Similarity-based methods that do not require knowledge of protein structures are particularly suitable for druggable genome-wide predictions of DPIs. We review 35 high-impact similarity-based predictors that were published in the past decade. We group them based on three types of similarities and their combinations that they use. We discuss and compare key aspects of these methods including source databases, internal databases and their predictive models. Using our novel benchmark database, we perform comparative empirical analysis of predictive performance of seven types of representative predictors that utilize each type of similarity individually and all possible combinations of similarities. We assess predictive quality at the database-wide DPI level and we are the first to also include evaluation over individual drugs. Our comprehensive analysis shows that predictors that use more similarity types outperform methods that employ fewer similarities, and that the model combining all three types of similarities secures area under the receiver operating characteristic curve of 0.93. We offer a comprehensive analysis of sensitivity of predictive performance to intrinsic and extrinsic characteristics of the considered predictors. We find that predictive performance is sensitive to low levels of similarities between sequences of the drug targets and several extrinsic properties of the input drug structures, drug profiles and drug targets. The benchmark database and a webserver for the seven predictors are freely available at http://biomine.cs.vcu.edu/servers/CONNECTOR/.

Master Regulators Connectivity Map: A Transcription Factors-Centered Approach to Drug Repositioning

Drug discovery is a very expensive and time-consuming endeavor. Fortunately, recent omics technologies and Systems Biology approaches introduced interesting new tools to achieve this task, facilitating the repurposing of already known drugs to new therapeutic assignments using gene expression data and bioinformatics. The inherent role of transcription factors in gene expression modulation makes them strong candidates for master regulators of phenotypic transitions. However, transcription factors expression itself usually does not reflect its activity changes due to post-transcriptional modifications and other complications. In this aspect, the use of high-throughput transcriptomic data may be employed to infer transcription factors-targets interactions and assess their activity through co-expression networks, which can be further used to search for drugs capable of reverting the gene expression profile of pathological phenotypes employing the connectivity maps paradigm. Following this idea, we argue that a module-oriented connectivity map approach using transcription factors-centered networks would aid the query for new repositioning candidates. Through a brief case study, we explored this idea in bipolar disorder, retrieving known drugs used in the usual clinical scenario as well as new candidates with potential therapeutic application in this disease. Indeed, the results of the case study indicate just how promising our approach may be to drug repositioning.

Querying of Disparate Association and Interaction Data in Biomedical Applications

In biomedical applications, network models are commonly used to represent interactions and higher-level associations among biological entities. Integrated analyses of these interaction and association data has proven useful in extracting knowledge, and generating novel hypotheses for biomedical research. However, since most datasets provide their own schema and query interface, opportunities for exploratory and integrative querying of disparate data are currently limited. In this study, we utilize RDF-based representations of biomedical interaction and association data to develop a querying framework that enables flexible specification and efficient processing of graph template matching queries. The proposed framework enables integrative querying of biomedical databases to discover complex patterns of associations among a diverse range of biological entities, including biomolecules, biological processes, organisms, and phenotypes. Our experimental results on the UniProt dataset show that the proposed framework can be used to efficiently process complex queries, and identify biologically relevant patterns of associations that cannot be readily obtained by querying each dataset independently.

Computational systems biology approaches for Parkinson's disease

Parkinson's disease (PD) is a prime example of a complex and heterogeneous disorder, characterized by multifaceted and varied motor- and non-motor symptoms and different possible interplays of genetic and environmental risk factors. While investigations of individual PD-causing mutations and risk factors in isolation are providing important insights to improve our understanding of the molecular mechanisms behind PD, there is a growing consensus that a more complete understanding of these mechanisms will require an integrative modeling of multifactorial disease-associated perturbations in molecular networks. Identifying and interpreting the combinatorial effects of multiple PD-associated molecular changes may pave the way towards an earlier and reliable diagnosis and more effective therapeutic interventions. This review provides an overview of computational systems biology approaches developed in recent years to study multifactorial molecular alterations in complex disorders, with a focus on PD research applications. Strengths and weaknesses of different cellular pathway and network analyses, and multivariate machine learning techniques for investigating PD-related omics data are discussed, and strategies proposed to exploit the synergies of multiple biological knowledge and data sources. A final outlook provides an overview of specific challenges and possible next steps for translating systems biology findings in PD to new omics-based diagnostic tools and targeted, drug-based therapeutic approaches.

Changing Trends in Computational Drug Repositioning

Efforts to maximize the indications potential and revenue from drugs that are already marketed are largely motivated by what Sir James Black, a Nobel Prize-winning pharmacologist advocated-"The most fruitful basis for the discovery of a new drug is to start with an old drug". However, rational design of drug mixtures poses formidable challenges because of the lack of or limited information about in vivo cell regulation, mechanisms of genetic pathway activation, and in vivo pathway interactions. Hence, most of the successfully repositioned drugs are the result of "serendipity", discovered during late phase clinical studies of unexpected but beneficial findings. The connections between drug candidates and their potential adverse drug reactions or new applications are often difficult to foresee because the underlying mechanism associating them is largely unknown, complex, or dispersed and buried in silos of information. Discovery of such multi-domain pharmacomodules-pharmacologically relevant sub-networks of biomolecules and/or pathways-from collection of databases by independent/simultaneous mining of multiple datasets is an active area of research. Here, while presenting some of the promising bioinformatics approaches and pipelines, we summarize and discuss the current and evolving landscape of computational drug repositioning.

Exploiting semantic patterns over biomedical knowledge graphs for predicting treatment and causative relations

BACKGROUND

Identifying new potential treatment options for medical conditions that cause human disease burden is a central task of biomedical research. Since all candidate drugs cannot be tested with animal and clinical trials, in vitro approaches are first attempted to identify promising candidates. Likewise, identifying different causal relations between biomedical entities is also critical to understand biomedical processes. Generally, natural language processing (NLP) and machine learning are used to predict specific relations between any given pair of entities using the distant supervision approach.

OBJECTIVE

To build high accuracy supervised predictive models to predict previously unknown treatment and causative relations between biomedical entities based only on semantic graph pattern features extracted from biomedical knowledge graphs.

METHODS

We used 7000 treats and 2918 causes hand-curated relations from the UMLS Metathesaurus to train and test our models. Our graph pattern features are extracted from simple paths connecting biomedical entities in the SemMedDB graph (based on the well-known SemMedDB database made available by the U.S. National Library of Medicine). Using these graph patterns connecting biomedical entities as features of logistic regression and decision tree models, we computed mean performance measures (precision, recall, F-score) over 100 distinct 80-20% train-test splits of the datasets. For all experiments, we used a positive:negative class imbalance of 1:10 in the test set to model relatively more realistic scenarios.

RESULTS

Our models predict treats and causes relations with high F-scores of 99% and 90% respectively. Logistic regression model coefficients also help us identify highly discriminative patterns that have an intuitive interpretation. We are also able to predict some new plausible relations based on false positives that our models scored highly based on our collaborations with two physician co-authors. Finally, our decision tree models are able to retrieve over 50% of treatment relations from a recently created external dataset.

CONCLUSIONS

We employed semantic graph patterns connecting pairs of candidate biomedical entities in a knowledge graph as features to predict treatment/causative relations between them. We provide what we believe is the first evidence in direct prediction of biomedical relations based on graph features. Our work complements lexical pattern based approaches in that the graph patterns can be used as additional features for weakly supervised relation prediction.

A Computational Systems Biology Approach for Identifying Candidate Drugs for Repositioning for Cardiovascular Disease

We report an in silico method to screen for receptors or pathways that could be targeted to elicit beneficial transcriptional changes in a cellular model of a disease of interest. In our method, we integrate: (1) a dataset of transcriptome responses of a cell line to a panel of drugs; (2) two sets of genes for the disease; and (3) mappings between drugs and the receptors or pathways that they target. We carried out a gene set enrichment analysis (GSEA) test for each of the two gene sets against a list of genes ordered by fold-change in response to a drug in a relevant cell line (HL60), with the overall score for a drug being the difference of the two enrichment scores. Next, we applied GSEA for drug targets based on drugs that have been ranked by their differential enrichment scores. The method ranks drugs by the degree of anti-correlation of their gene-level transcriptional effects on the cell line with the genes in the disease gene sets. We applied the method to data from (1) CMap 2.0; (2) gene sets from two transcriptome profiling studies of atherosclerosis; and (3) a combined dataset of drug/target information. Our analysis recapitulated known targets related to CVD (e.g., PPARγ; HMG-CoA reductase, HDACs) and novel targets (e.g., amine oxidase A, δ-opioid receptor). We conclude that combining disease-associated gene sets, drug-transcriptome-responses datasets and drug-target annotations can potentially be useful as a screening tool for diseases that lack an accepted cellular model for in vitro screening.

Connecting genetics and gene expression data for target prioritisation and drug repositioning

Developing new drugs continues to be a highly inefficient and costly business. By repurposing an existing compound for a different indication, drug repositioning offers an attractive alternative to traditional drug discovery. Most of these approaches work by matching transcriptional disease signatures to anti-correlated gene expression profiles of drug perturbations. Genome-wide association studies (GWASs) are of great interest to researchers in the pharmaceutical industry because drug programmes with supporting genetic evidence are more likely to successfully progress through the drug discovery pipeline. Here, we present a systematic approach to generate drug repositioning hypothesis based on disease genetics by mining public repositories of GWAS data and drug transcriptomic profiles. We find that genes genetically associated with a certain disease are more likely to be differentially expressed in the same disease (p-value = 1.54e-17 and AUC = 0.75) and that, in existing drug - disease combinations, genes significantly up- or down-regulated after drug treatment are enriched for genes genetically associated with that disease (p-value = 1.1e-79 and AUC = 0.64). Finally, we use this framework to generate and rank novel GWAS-driven drug repositioning predictions.

Reverse Screening Methods to Search for the Protein Targets of Chemopreventive Compounds

This article is a systematic review of reverse screening methods used to search for the protein targets of chemopreventive compounds or drugs. Typical chemopreventive compounds include components of traditional Chinese medicine, natural compounds and Food and Drug Administration (FDA)-approved drugs. Such compounds are somewhat selective but are predisposed to bind multiple protein targets distributed throughout diverse signaling pathways in human cells. In contrast to conventional virtual screening, which identifies the ligands of a targeted protein from a compound database, reverse screening is used to identify the potential targets or unintended targets of a given compound from a large number of receptors by examining their known ligands or crystal structures. This method, also known as in silico or computational target fishing, is highly valuable for discovering the target receptors of query molecules from terrestrial or marine natural products, exploring the molecular mechanisms of chemopreventive compounds, finding alternative indications of existing drugs by drug repositioning, and detecting adverse drug reactions and drug toxicity. Reverse screening can be divided into three major groups: shape screening, pharmacophore screening and reverse docking. Several large software packages, such as Schrödinger and Discovery Studio; typical software/network services such as ChemMapper, PharmMapper, idTarget, and INVDOCK; and practical databases of known target ligands and receptor crystal structures, such as ChEMBL, BindingDB, and the Protein Data Bank (PDB), are available for use in these computational methods. Different programs, online services and databases have different applications and constraints. Here, we conducted a systematic analysis and multilevel classification of the computational programs, online services and compound libraries available for shape screening, pharmacophore screening and reverse docking to enable non-specialist users to quickly learn and grasp the types of calculations used in protein target fishing. In addition, we review the main features of these methods, programs and databases and provide a variety of examples illustrating the application of one or a combination of reverse screening methods for accurate target prediction.

A systems biology approach to predict and characterize human gut microbial metabolites in colorectal cancer

Colorectal cancer (CRC) is the second leading cause of cancer-related deaths. It is estimated that about half the cases of CRC occurring today are preventable. Recent studies showed that human gut microbiota and their collective metabolic outputs play important roles in CRC. However, the mechanisms by which human gut microbial metabolites interact with host genetics in contributing CRC remain largely unknown. We hypothesize that computational approaches that integrate and analyze vast amounts of publicly available biomedical data have great potential in better understanding how human gut microbial metabolites are mechanistically involved in CRC. Leveraging vast amount of publicly available data, we developed a computational algorithm to predict human gut microbial metabolites for CRC. We validated the prediction algorithm by showing that previously known CRC-associated gut microbial metabolites ranked highly (mean ranking: top 10.52%; median ranking: 6.29%; p-value: 3.85E-16). Moreover, we identified new gut microbial metabolites likely associated with CRC. Through computational analysis, we propose potential roles for tartaric acid, the top one ranked metabolite, in CRC etiology. In summary, our data-driven computation-based study generated a large amount of associations that could serve as a starting point for further experiments to refute or validate these microbial metabolite associations in CRC cancer.

In silico cancer research towards 3R

BACKGROUND

Improving our understanding of cancer and other complex diseases requires integrating diverse data sets and algorithms. Intertwining in vivo and in vitro data and in silico models are paramount to overcome intrinsic difficulties given by data complexity. Importantly, this approach also helps to uncover underlying molecular mechanisms. Over the years, research has introduced multiple biochemical and computational methods to study the disease, many of which require animal experiments. However, modeling systems and the comparison of cellular processes in both eukaryotes and prokaryotes help to understand specific aspects of uncontrolled cell growth, eventually leading to improved planning of future experiments. According to the principles for humane techniques milestones in alternative animal testing involve in vitro methods such as cell-based models and microfluidic chips, as well as clinical tests of microdosing and imaging. Up-to-date, the range of alternative methods has expanded towards computational approaches, based on the use of information from past in vitro and in vivo experiments. In fact, in silico techniques are often underrated but can be vital to understanding fundamental processes in cancer. They can rival accuracy of biological assays, and they can provide essential focus and direction to reduce experimental cost.

MAIN BODY

We give an overview on in vivo, in vitro and in silico methods used in cancer research. Common models as cell-lines, xenografts, or genetically modified rodents reflect relevant pathological processes to a different degree, but can not replicate the full spectrum of human disease. There is an increasing importance of computational biology, advancing from the task of assisting biological analysis with network biology approaches as the basis for understanding a cell's functional organization up to model building for predictive systems.

CONCLUSION

Underlining and extending the in silico approach with respect to the 3Rs for replacement, reduction and refinement will lead cancer research towards efficient and effective precision medicine. Therefore, we suggest refined translational models and testing methods based on integrative analyses and the incorporation of computational biology within cancer research.

A two-tiered unsupervised clustering approach for drug repositioning through heterogeneous data integration

Background

Drug repositioning is the process of identifying new uses for existing drugs. Computational drug repositioning methods can reduce the time, costs and risks of drug development by automating the analysis of the relationships in pharmacology networks. Pharmacology networks are large and heterogeneous. Clustering drugs into small groups can simplify large pharmacology networks, these subgroups can also be used as a starting point for repositioning drugs. In this paper, we propose a two-tiered drug-centric unsupervised clustering approach for drug repositioning, integrating heterogeneous drug data profiles: drug-chemical, drug-disease, drug-gene, drug-protein and drug-side effect relationships.

Results

The proposed drug repositioning approach is threefold; (i) clustering drugs based on their homogeneous profiles using the Growing Self Organizing Map (GSOM); (ii) clustering drugs based on drug-drug relation matrices based on the previous step, considering three state-of-the-art graph clustering methods; and (iii) inferring drug repositioning candidates and assigning a confidence value for each identified candidate. In this paper, we compare our two-tiered clustering approach against two existing heterogeneous data integration approaches with reference to the Anatomical Therapeutic Chemical (ATC) classification, using GSOM. Our approach yields Normalized Mutual Information (NMI) and Standardized Mutual Information (SMI) of 0.66 and 36.11, respectively, while the two existing methods yield NMI of 0.60 and 0.64 and SMI of 22.26 and 33.59. Moreover, the two existing approaches failed to produce useful cluster separations when using graph clustering algorithms while our approach is able to identify useful clusters for drug repositioning. Furthermore, we provide clinical evidence for four predicted results (Chlorthalidone, Indomethacin, Metformin and Thioridazine) to support that our proposed approach can be reliably used to infer ATC code and drug repositioning.

Conclusion

The proposed two-tiered unsupervised clustering approach is suitable for drug clustering and enables heterogeneous data integration. It also enables identifying reliable repositioning drug candidates with reference to ATC therapeutic classification. The repositioning drug candidates identified consistently by multiple clustering algorithms and with high confidence have a higher possibility of being effective repositioning candidates.

Electronic supplementary material

The online version of this article (10.1186/s12859-018-2123-4) contains supplementary material, which is available to authorized users.

Computational repositioning and preclinical validation of mifepristone for human vestibular schwannoma

The computational repositioning of existing drugs represents an appealing avenue for identifying effective compounds to treat diseases with no FDA-approved pharmacotherapies. Here we present the largest meta-analysis to date of differential gene expression in human vestibular schwannoma (VS), a debilitating intracranial tumor, and use these data to inform the first application of algorithm-based drug repositioning for this tumor class. We apply an open-source computational drug repositioning platform to gene expression data from 80 patient tumors and identify eight promising FDA-approved drugs with potential for repurposing in VS. Of these eight, mifepristone, a progesterone and glucocorticoid receptor antagonist, consistently and adversely affects the morphology, metabolic activity, and proliferation of primary human VS cells and HEI-193 human schwannoma cells. Mifepristone treatment reduces VS cell viability more significantly than cells derived from patient meningiomas, while healthy human Schwann cells remain unaffected. Our data recommend a Phase II clinical trial of mifepristone in VS.

Opportunities and obstacles for deep learning in biology and medicine

Deep learning describes a class of machine learning algorithms that are capable of combining raw inputs into layers of intermediate features. These algorithms have recently shown impressive results across a variety of domains. Biology and medicine are data-rich disciplines, but the data are complex and often ill-understood. Hence, deep learning techniques may be particularly well suited to solve problems of these fields. We examine applications of deep learning to a variety of biomedical problems-patient classification, fundamental biological processes and treatment of patients-and discuss whether deep learning will be able to transform these tasks or if the biomedical sphere poses unique challenges. Following from an extensive literature review, we find that deep learning has yet to revolutionize biomedicine or definitively resolve any of the most pressing challenges in the field, but promising advances have been made on the prior state of the art. Even though improvements over previous baselines have been modest in general, the recent progress indicates that deep learning methods will provide valuable means for speeding up or aiding human investigation. Though progress has been made linking a specific neural network's prediction to input features, understanding how users should interpret these models to make testable hypotheses about the system under study remains an open challenge. Furthermore, the limited amount of labelled data for training presents problems in some domains, as do legal and privacy constraints on work with sensitive health records. Nonetheless, we foresee deep learning enabling changes at both bench and bedside with the potential to transform several areas of biology and medicine.

Personalised drug repositioning for Clear Cell Renal Cell Carcinoma using gene expression

Reversal of cancer gene expression is predictive of therapeutic potential and can be used to find new indications for existing drugs (drug repositioning). Gene expression reversal potential is currently calculated, in almost all studies, by pre-aggregating all tumour samples into a single group signature or a limited number of molecular subtype signatures. Here, we investigate whether drug repositioning based on individual tumour sample gene expression signatures outperforms the use of tumour group and subtype signatures. The tumour signatures were created using 534 tumour samples and 72 matched normal samples from 530 clear cell renal cell carcinoma (ccRCC) patients. More than 20,000 drug signatures were extracted from the CMAP and LINCS databases. We show that negative enrichment of individual tumour samples correlated (Spearman's rho = 0.15) much better with the amount of differentially expressed genes in drug signatures than with the tumour group signature (Rho = 0.08) and the 4 tumour subtype signatures (Rho 0.036-0.11). Targeted drugs used against ccRCC, such as sirolimus and temsirolimus, which could not be identified with the pre-aggregated tumour signatures could be recovered using individual sample analysis. Thus, drug repositioning can be personalized by taking into account the gene expression profile of the individual's tumour sample.

Large-scale computational drug repositioning to find treatments for rare diseases

Rare, or orphan, diseases are conditions afflicting a small subset of people in a population. Although these disorders collectively pose significant health care problems, drug companies require government incentives to develop drugs for rare diseases due to extremely limited individual markets. Computer-aided drug repositioning, i.e., finding new indications for existing drugs, is a cheaper and faster alternative to traditional drug discovery offering a promising venue for orphan drug research. Structure-based matching of drug-binding pockets is among the most promising computational techniques to inform drug repositioning. In order to find new targets for known drugs ultimately leading to drug repositioning, we recently developed eMatchSite, a new computer program to compare drug-binding sites. In this study, eMatchSite is combined with virtual screening to systematically explore opportunities to reposition known drugs to proteins associated with rare diseases. The effectiveness of this integrated approach is demonstrated for a kinase inhibitor, which is a confirmed candidate for repositioning to synapsin Ia. The resulting dataset comprises 31,142 putative drug-target complexes linked to 980 orphan diseases. The modeling accuracy is evaluated against the structural data recently released for tyrosine-protein kinase HCK. To illustrate how potential therapeutics for rare diseases can be identified, we discuss a possibility to repurpose a steroidal aromatase inhibitor to treat Niemann-Pick disease type C. Overall, the exhaustive exploration of the drug repositioning space exposes new opportunities to combat orphan diseases with existing drugs. DrugBank/Orphanet repositioning data are freely available to research community at https://osf.io/qdjup/.

Big Data in Drug Discovery

Interpretation of Big Data in the drug discovery community should enhance project timelines and reduce clinical attrition through improved early decision making. The issues we encounter start with the sheer volume of data and how we first ingest it before building an infrastructure to house it to make use of the data in an efficient and productive way. There are many problems associated with the data itself including general reproducibility, but often, it is the context surrounding an experiment that is critical to success. Help, in the form of artificial intelligence (AI), is required to understand and translate the context. On the back of natural language processing pipelines, AI is also used to prospectively generate new hypotheses by linking data together. We explain Big Data from the context of biology, chemistry and clinical trials, showcasing some of the impressive public domain sources and initiatives now available for interrogation.

Computational prediction of drug–target interactions using chemogenomic approaches: an empirical survey

Computational prediction of drug–target interactions (DTIs) has become an essential task in the drug discovery process. It narrows down the search space for interactions by suggesting potential interaction candidates for validation via wet-lab experiments that are well known to be expensive and time-consuming. In this article, we aim to provide a comprehensive overview and empirical evaluation on the computational DTI prediction techniques, to act as a guide and reference for our fellow researchers. Specifically, we first describe the data used in such computational DTI prediction efforts. We then categorize and elaborate the state-of-the-art methods for predicting DTIs. Next, an empirical comparison is performed to demonstrate the prediction performance of some representative methods under different scenarios. We also present interesting findings from our evaluation study, discussing the advantages and disadvantages of each method. Finally, we highlight potential avenues for further enhancement of DTI prediction performance as well as related research directions.

Translating Knowledge Into Therapy for Acute Kidney Injury

No therapies have been shown to improve outcomes in patients with acute kidney injury (AKI). Given the high morbidity and mortality associated with AKI this represents an important unmet medical need. A common feature of all of the therapeutic development efforts for AKI is that none were driven by target selection or preclinical modeling that was based primarily on human data. This is important when considering a heterogeneous and dynamic condition such as AKI, in which in the absence of more accurate molecular classifications, clinical cohorts are likely to include patients with different types of injury at different stages in the injury and repair continuum. The National Institutes of Health precision medicine initiative offers an opportunity to address this. By creating a molecular tissue atlas of AKI, defining patient subgroups, and identifying critical cells and pathways involved in human AKI, this initiative has the potential to transform our current approach to therapeutic discovery. In this review, we discuss the opportunities and challenges that this initiative presents, with a specific focus on AKI, what additional efforts will be needed to apply these discoveries to therapeutic development, and how we believe this effort might lead to the development of new therapeutics for subsets of patients with AKI.