Adverse Drug Reaction Prediction Using Scores Produced by Large-Scale Drug-Protein Target Docking on High-Performance Computing Machines

Late-stage or post-market identification of adverse drug reactions (ADRs) is a significant public health issue and a source of major economic liability for drug development. Thus, reliable in silico screening of drug candidates for possible ADRs would be advantageous. In this work, we introduce a computational approach that predicts ADRs by combining the results of molecular docking and leverages known ADR information from DrugBank and SIDER. We employed a recently parallelized version of AutoDock Vina (VinaLC) to dock 906 small molecule drugs to a virtual panel of 409 DrugBank protein targets. L1-regularized logistic regression models were trained on the resulting docking scores of a 560 compound subset from the initial 906 compounds to predict 85 side effects, grouped into 10 ADR phenotype groups. Only 21% (87 out of 409) of the drug-protein binding features involve known targets of the drug subset, providing a significant probe of off-target effects. As a control, associations of this drug subset with the 555 annotated targets of these compounds, as reported in DrugBank, were used as features to train a separate group of models. The Vina off-target models and the DrugBank on-target models yielded comparable median area-under-the-receiver-operating-characteristic-curves (AUCs) during 10-fold cross-validation (0.60–0.69 and 0.61–0.74, respectively). Evidence was found in the PubMed literature to support several putative ADR-protein associations identified by our analysis. Among them, several associations between neoplasm-related ADRs and known tumor suppressor and tumor invasiveness marker proteins were found. A dual role for interstitial collagenase in both neoplasms and aneurysm formation was also identified. These associations all involve off-target proteins and could not have been found using available drug/on-target interaction data. This study illustrates a path forward to comprehensive ADR virtual screening that can potentially scale with increasing number of CPUs to tens of thousands of protein targets and millions of potential drug candidates.     

Gene Ontology and KEGG Pathway Enrichment Analysis of a Drug Target-Based Classification System

Drug-target interaction (DTI) is a key aspect in pharmaceutical research. With the ever-increasing new drug data resources, computational approaches have emerged as powerful and labor-saving tools in predicting new DTIs. However, so far, most of these predictions have been based on structural similarities rather than biological relevance. In this study, we proposed for the first time a “GO and KEGG enrichment score” method to represent a certain category of drug molecules by further classification and interpretation of the DTI database. A benchmark dataset consisting of 2,015 drugs that are assigned to nine categories ((1) G protein-coupled receptors, (2) cytokine receptors, (3) nuclear receptors, (4) ion channels, (5) transporters, (6) enzymes, (7) protein kinases, (8) cellular antigens and (9) pathogens) was constructed by collecting data from KEGG. We analyzed each category and each drug for its contribution in GO terms and KEGG pathways using the popular feature selection “minimum redundancy maximum relevance (mRMR)” method, and key GO terms and KEGG pathways were extracted. Our analysis revealed the top enriched GO terms and KEGG pathways of each drug category, which were highly enriched in the literature and clinical trials. Our results provide for the first time the biological relevance among drugs, targets and biological functions, which serves as a new basis for future DTI predictions.     

A new methodology to associate SNPs with human diseases according to their pathway related context

Genome-wide association studies (GWAS) with hundreds of żthousands of single nucleotide polymorphisms (SNPs) are popular strategies to reveal the genetic basis of human complex diseases. Despite many successes of GWAS, it is well recognized that new analytical approaches have to be integrated to achieve their full potential. Starting with a list of SNPs, found to be associated with disease in GWAS, here we propose a novel methodology to devise functionally important KEGG pathways through the identification of genes within these pathways, where these genes are obtained from SNP analysis. Our methodology is based on functionalization of important SNPs to identify effected genes and disease related pathways. We have tested our methodology on WTCCC Rheumatoid Arthritis (RA) dataset and identified: i) previously known RA related KEGG pathways (e.g., Toll-like receptor signaling, Jak-STAT signaling, Antigen processing, Leukocyte transendothelial migration and MAPK signaling pathways); ii) additional KEGG pathways (e.g., Pathways in cancer, Neurotrophin signaling, Chemokine signaling pathways) as associated with RA. Furthermore, these newly found pathways included genes which are targets of RA-specific drugs. Even though GWAS analysis identifies 14 out of 83 of those drug target genes; newly found functionally important KEGG pathways led to the discovery of 25 out of 83 genes, known to be used as drug targets for the treatment of RA. Among the previously known pathways, we identified additional genes associated with RA (e.g. Antigen processing and presentation, Tight junction). Importantly, within these pathways, the associations between some of these additionally found genes, such as HLA-C, HLA-G, PRKCQ, PRKCZ, TAP1, TAP2 and RA were verified by either OMIM database or by literature retrieved from the NCBI PubMed module. With the whole-genome sequencing on the horizon, we show that the full potential of GWAS can be achieved by integrating pathway and network-oriented analysis and prior knowledge from functional properties of a SNP.     

Bias in Spontaneous Reporting of Adverse Drug Reactions in Japan

Background

Attitudes of healthcare professionals regarding spontaneous reporting of adverse drug reactions (ADRs) in Japan are not well known, and Japan’s unique system of surveillance, called early post-marketing phase vigilance (EPPV), may affect these reporting attitudes. Our objectives were to describe potential effects of EPPV and to test whether ADR seriousness, prominence, and frequency are related to changes in reporting over time.

Methods

A manufacturer’s database of spontaneous ADR reports was used to extract data from individual case safety reports for 5 drugs subject to EPPV. The trend of reporting and the time lag between ADR onset and reporting to the manufacturer were examined. The following indices for ADRs occurring with each drug were calculated and analyzed to assess reporting trends: Serious:Non-serious ratio, High prominence:Low prominence ratio, and High frequency:Low frequency ratio.

Results

For all 5 drugs, the time lag between ADR onset and reporting to the manufacturer was shorter in the EPPV period than in the post-EPPV period. All drugs showed higher Serious:Non-serious ratios in the post-EPPV period. No specific patterns were observed for the High prominence:Low prominence ratio. The High frequency:Low frequency ratio for peginterferon alpha-2a and sevelamer hydrochloride decreased steadily throughout the study period.

Conclusions

Healthcare professionals may be more likely to report serious ADRs than to report non-serious ADRs, but the effect of event prominence on reporting trends is still unclear. Factors associated with ADR reporting attitude in Japan might be different from those in other countries because of EPPV and the involvement of medical representatives in the spontaneous reporting process. Pharmacovigilance specialists should therefore be cautious when comparing data between different time periods or different countries. Further studies are needed to elucidate the underlying mechanism of spontaneous ADR reporting in Japan.     

熵权法在药品不良反应信号筛选及综合风险评价中的应用

目的：在运用经典方法对药品不良反应信号进行检测后,对信号进行再筛选并评价药品的综合风险。方法：以江苏省药品不良反应监测网络数据库为资料来源,SQL server 为后台数据库,Matlab 为算法主要实现工具,运用熵权法结合专家评分,对BCPNN 方法检测出的信号进行再调整和评级。结果：运用综合权重进行调整后,发现不同药品不良反应信号的强弱发生了一定的变化。其中,氟喹诺酮类药品中的加替沙星导致低血糖、呼吸困难等药品不良反应的风险较信号评级之前有所增加,需要专家在评审时更为留意。结论：将熵权法运用在药品不良反应信号的监测中,可使信号更接近客观筛选和主观判断的平衡值,部分罕发但严重的药品不良反应信号得到发现和重视,并能方便地研究药物引起的多种药品不良反应的综合风险。     

Large-scale elucidation of drug response pathways in humans

Elucidating signaling pathways is a fundamental step in understanding cellular processes and developing new therapeutic strategies. Here we introduce a method for the large-scale elucidation of signaling pathways involved in cellular response to drugs. Combining drug targets, drug response expression profiles, and the human physical interaction network, we infer 99 human drug response pathways and study their properties. Based on the newly inferred pathways, we develop a pathway-based drug-drug similarity measure and compare it to two common, gold standard drug-drug similarity measures. Remarkably, our measure provides better correspondence to these gold standards than similarity measures that are based on associations between drugs and known pathways, or on drug-specific gene expression profiles. It further improves the prediction of drug side effects and indications, elucidating specific response pathways that may be associated with these drug properties. Supplementary Material for this article is available at www.liebertonline.com/cmb.     

TRMP: a database of therapeutically relevant multiple pathways

SUMMARY: Disease processes often involve crosstalks between proteins in different pathways. Different proteins have been used as separate therapeutic targets for the same disease. Synergetic targeting of multiple targets has been explored in combination therapy of a number of diseases. Potential harmful interactions of multiple targeting have also been closely studied. To facilitate mechanistic study of drug actions and a more comprehensive understanding the relationship between different targets of the same disease, it is useful to develop a database of known therapeutically relevant multiple pathways (TRMPs). Information about non-target proteins and natural small molecules involved in these pathways also provides useful hint for searching new therapeutic targets and facilitate the understanding of how therapeutic targets interact with other molecules in performing specific tasks. The TRMPs database is designed to provide information about such multiple pathways along with related therapeutic targets, corresponding drugs/ligands, targeted disease conditions, constituent individual pathways, structural and functional information about each protein in the pathways. Cross links to other databases are also introduced to facilitate the access of information about individual pathways and proteins. AVAILABILITY: This database can be accessed at http://bidd.nus.edu.sg/group/trmp/trmp.asp and it currently contains 11 entries of multiple pathways, 97 entries of individual pathways, 120 targets covering 72 disease conditions together with 120 sets of drugs directed at each of these targets. Each entry can be retrieved through multiple methods including multiple pathway name, individual pathway name and disease name. SUPPLEMENTARY INFORMATION: http://bidd.nus.edu.sg/group/trmp/sm.pdf     

Co-clustering phenome�Cgenome for phenotype classification and disease gene discovery

Understanding the categorization of human diseases is critical for reliably identifying disease causal genes. Recently, genome-wide studies of abnormal chromosomal locations related to diseases have mapped >2000 phenotype–gene relations, which provide valuable information for classifying diseases and identifying candidate genes as drug targets. In this article, a regularized non-negative matrix tri-factorization (R-NMTF) algorithm is introduced to co-cluster phenotypes and genes, and simultaneously detect associations between the detected phenotype clusters and gene clusters. The R-NMTF algorithm factorizes the phenotype–gene association matrix under the prior knowledge from phenotype similarity network and protein–protein interaction network, supervised by the label information from known disease classes and biological pathways. In the experiments on disease phenotype–gene associations in OMIM and KEGG disease pathways, R-NMTF significantly improved the classification of disease phenotypes and disease pathway genes compared with support vector machines and Label Propagation in cross-validation on the annotated phenotypes and genes. The newly predicted phenotypes in each disease class are highly consistent with human phenotype ontology annotations. The roles of the new member genes in the disease pathways are examined and validated in the protein–protein interaction subnetworks. Extensive literature review also confirmed many new members of the disease classes and pathways as well as the predicted associations between disease phenotype classes and pathways.     

High-throughput framework for genetic analyses of adverse drug reactions using electronic health records

Understanding the contribution of genetic variation to drug response can improve the delivery of precision medicine. However, genome-wide association studies (GWAS) for drug response are uncommon and are often hindered by small sample sizes. We present a high-throughput framework to efficiently identify eligible patients for genetic studies of adverse drug reactions (ADRs) using “drug allergy” labels from electronic health records (EHRs). As a proof-of-concept, we conducted GWAS for ADRs to 14 common drug/drug groups with 81,739 individuals from Vanderbilt University Medical Center’s BioVU DNA Biobank. We identified 7 genetic loci associated with ADRs at P < 5 × 10⁻⁸, including known genetic associations such as CYP2D6 and OPRM1 for CYP2D6-metabolized opioid ADR. Additional expression quantitative trait loci and phenome-wide association analyses added evidence to the observed associations. Our high-throughput framework is both scalable and portable, enabling impactful pharmacogenomic research to improve precision medicine.     

Translating clinical findings into knowledge in drug safety evaluation--drug induced liver injury prediction system (DILIps)

Drug-induced liver injury (DILI) is a significant concern in drug development due to the poor concordance between preclinical and clinical findings of liver toxicity. We hypothesized that the DILI types (hepatotoxic side effects) seen in the clinic can be translated into the development of predictive in silico models for use in the drug discovery phase. We identified 13 hepatotoxic side effects with high accuracy for classifying marketed drugs for their DILI potential. We then developed in silico predictive models for each of these 13 side effects, which were further combined to construct a DILI prediction system (DILIps). The DILIps yielded 60-70% prediction accuracy for three independent validation sets. To enhance the confidence for identification of drugs that cause severe DILI in humans, the "Rule of Three" was developed in DILIps by using a consensus strategy based on 13 models. This gave high positive predictive value (91%) when applied to an external dataset containing 206 drugs from three independent literature datasets. Using the DILIps, we screened all the drugs in DrugBank and investigated their DILI potential in terms of protein targets and therapeutic categories through network modeling. We demonstrated that two therapeutic categories, anti-infectives for systemic use and musculoskeletal system drugs, were enriched for DILI, which is consistent with current knowledge. We also identified protein targets and pathways that are related to drugs that cause DILI by using pathway analysis and co-occurrence text mining. While marketed drugs were the focus of this study, the DILIps has a potential as an evaluation tool to screen and prioritize new drug candidates or chemicals, such as environmental chemicals, to avoid those that might cause liver toxicity. We expect that the methodology can be also applied to other drug safety endpoints, such as renal or cardiovascular toxicity.     

miRNA signature associated with outcome of gastric cancer patients following chemotherapy

Background

Resistance to chemotherapy severely limits the effectiveness of chemotherapy drugs in treating cancer. Still, the mechanisms and critical pathways that contribute to chemotherapy resistance are relatively unknown. This study elucidates the chemoresistance-associated pathways retrieved from the integrated biological interaction networks and identifies signature genes relevant for chemotherapy resistance.

Methods

An integrated network was constructed by collecting multiple metabolic interactions from public databases and the k-shortest path algorithm was implemented to identify chemoresistant related pathways. The identified pathways were then scored using differential expression values from microarray data in chemosensitive and chemoresistant ovarian and lung cancers. Finally, another pathway database, Reactome, was used to evaluate the significance of genes within each filtered pathway based on topological characteristics.

Results

By this method, we discovered pathways specific to chemoresistance. Many of these pathways were consistent with or supported by known involvement in chemotherapy. Experimental results also indicated that integration of pathway structure information with gene differential expression analysis can identify dissimilar modes of gene reactions between chemosensitivity and chemoresistance. Several identified pathways can increase the development of chemotherapeutic resistance and the predicted signature genes are involved in drug resistant during chemotherapy. In particular, we observed that some genes were key factors for joining two or more metabolic pathways and passing down signals, which may be potential key targets for treatment.

Conclusions

This study is expected to identify targets for chemoresistant issues and highlights the interconnectivity of chemoresistant mechanisms. The experimental results not only offer insights into the mode of biological action of drug resistance but also provide information on potential key targets (new biological hypothesis) for further drug-development efforts.     

The transportome of the malaria parasite

Membrane transport proteins, also known as transporters, control the movement of ions, nutrients, metabolites, and waste products across the membranes of a cell and are central to its biology. Proteins of this type also serve as drug targets and are key players in the phenomenon of drug resistance. The malaria parasite has a relatively reduced transportome, with only approximately 2.5% of its genes encoding transporters. Even so, assigning functions and physiological roles to these proteins, and ascertaining their contributions to drug action and drug resistance, has been very challenging. This review presents a detailed critique and synthesis of the disruption phenotypes, protein subcellular localisations, protein functions (observed or predicted), and links to antimalarial drug resistance for each of the parasite's transporter genes. The breadth and depth of the gene disruption data are particularly impressive, with at least one phenotype determined in the parasite's asexual blood stage for each transporter gene, and multiple phenotypes available for 76% of the genes. Analysis of the curated data set revealed there to be relatively little redundancy in the Plasmodium transportome; almost two‐thirds of the parasite's transporter genes are essential or required for normal growth in the asexual blood stage of the parasite, and this proportion increased to 78% when the disruption phenotypes available for the other parasite life stages were included in the analysis. These observations, together with the finding that 22% of the transportome is implicated in the parasite's resistance to existing antimalarials and/or drugs within the development pipeline, indicate that transporters are likely to serve, or are already serving, as drug targets. Integration of the different biological and bioinformatic data sets also enabled the selection of candidates for transport processes known to be essential for parasite survival, but for which the underlying proteins have thus far remained undiscovered. These include potential transporters of pantothenate, isoleucine, or isopentenyl diphosphate, as well as putative anion‐selective channels that may serve as the pore component of the parasite's ‘new permeation pathways’. Other novel insights into the parasite's biology included the identification of transporters for the potential development of antimalarial treatments, transmission‐blocking drugs, prophylactics, and genetically attenuated vaccines. The syntheses presented herein set a foundation for elucidating the functions and physiological roles of key members of the Plasmodium transportome and, ultimately, to explore and realise their potential as therapeutic targets.     

Identification of common biological pathways and drug targets across multiple respiratory viruses based on human host gene expression analysis

Background

Pandemic and seasonal respiratory viruses are a major global health concern. Given the genetic diversity of respiratory viruses and the emergence of drug resistant strains, the targeted disruption of human host-virus interactions is a potential therapeutic strategy for treating multi-viral infections. The availability of large-scale genomic datasets focused on host-pathogen interactions can be used to discover novel drug targets as well as potential opportunities for drug repositioning.

Methods/Results

In this study, we performed a large-scale analysis of microarray datasets involving host response to infections by influenza A virus, respiratory syncytial virus, rhinovirus, SARS-coronavirus, metapneumonia virus, coxsackievirus and cytomegalovirus. Common genes and pathways were found through a rigorous, iterative analysis pipeline where relevant host mRNA expression datasets were identified, analyzed for quality and gene differential expression, then mapped to pathways for enrichment analysis. Possible repurposed drugs targets were found through database and literature searches. A total of 67 common biological pathways were identified among the seven different respiratory viruses analyzed, representing fifteen laboratories, nine different cell types, and seven different array platforms. A large overlap in the general immune response was observed among the top twenty of these 67 pathways, adding validation to our analysis strategy. Of the top five pathways, we found 53 differentially expressed genes affected by at least five of the seven viruses. We suggest five new therapeutic indications for existing small molecules or biological agents targeting proteins encoded by the genes F3, IL1B, TNF, CASP1 and MMP9. Pathway enrichment analysis also identified a potential novel host response, the Parkin-Ubiquitin Proteasomal System (Parkin-UPS) pathway, which is known to be involved in the progression of neurodegenerative Parkinson''s disease.

Conclusions

Our study suggests that multiple and diverse respiratory viruses invoke several common host response pathways. Further analysis of these pathways suggests potential opportunities for therapeutic intervention.     

Factors Affecting the Timing of Signal Detection of Adverse Drug Reactions

We investigated factors affecting the timing of signal detection by comparing variations in reporting time of known and unknown ADRs after initial drug release in the USA. Data on adverse event reactions (AERs) submitted to U.S. FDA was used. Six ADRs associated with 6 drugs (rosuvastatin, aripiprazole, teriparatide, telithromycin, exenatide, varenicline) were investigated: Changes in the proportional reporting ratio, reporting odds ratio, and information component as indexes of signal detection were followed every 3 months after each drugs release, and the time for detection of signals was investigated. The time for the detection of signal to be detected after drug release in the USA was 2–10 months for known ADRs and 19–44 months for unknown ones. The median lag time for known and unknown ADRs was 99.0–122.5 days and 185.5–306.0 days, respectively. When the FDA released advisory information on rare but potentially serious health risks of an unknown ADR, the time lag to report from the onset of ADRs to the FDA was shorter. This study suggested that one factor affecting signal detection time is whether an ADR was known or unknown at release.     

Study of drug function based on similarity of pathway fingerprint

Drugs sharing similar therapeutic function may not bind to the same group of targets. However, their targets may be involved in similar pathway profiles which are associated with certain pathological process. In this study, pathway fingerprint was introduced to indicate the profile of significant pathways being influenced by the targets of drugs. Then drug-drug network was further constructed based on significant similarity of pathway fingerprints. In this way, the functions of a drug may be hinted by the enriched therapeutic functions of its neighboring drugs. In the test of 911 FDA approved drugs with more than one known target, 471 drugs could be connected into networks. 760 significant associations of drug-therapeutic function were generated, among which around 60% of them were supported by scientific literatures or ATC codes of drug functional classification. Therefore, pathway fingerprints may be useful to further study on the potential function of known drugs, or the unknown function of new drugs.     

A Systematic Investigation of Computation Models for Predicting Adverse Drug Reactions (ADRs)

Background

Early and accurate identification of adverse drug reactions (ADRs) is critically important for drug development and clinical safety. Computer-aided prediction of ADRs has attracted increasing attention in recent years, and many computational models have been proposed. However, because of the lack of systematic analysis and comparison of the different computational models, there remain limitations in designing more effective algorithms and selecting more useful features. There is therefore an urgent need to review and analyze previous computation models to obtain general conclusions that can provide useful guidance to construct more effective computational models to predict ADRs.

Principal Findings

In the current study, the main work is to compare and analyze the performance of existing computational methods to predict ADRs, by implementing and evaluating additional algorithms that have been earlier used for predicting drug targets. Our results indicated that topological and intrinsic features were complementary to an extent and the Jaccard coefficient had an important and general effect on the prediction of drug-ADR associations. By comparing the structure of each algorithm, final formulas of these algorithms were all converted to linear model in form, based on this finding we propose a new algorithm called the general weighted profile method and it yielded the best overall performance among the algorithms investigated in this paper.

Conclusion

Several meaningful conclusions and useful findings regarding the prediction of ADRs are provided for selecting optimal features and algorithms.     

Errors in medication history at hospital admission: prevalence and predicting factors

Background

Data on adverse drug reactions (ADRs) related to antiretroviral (ARV) use in public health practice are few indicating the need for ART safety surveillance in clinical care.

Objectives

To evaluate the incidence, type and risk factors associated with adverse drug reactions (ADRs) among patients on antiretroviral drugs (ARV).

Methods

Patients initiated on ARVs between May 2006 and May 2009 were evaluated in a retrospective cohort analysis in three health facilities in Nigeria. Regimens prescribed include nucleoside backbone of zidovudine (AZT)/lamivudine (3TC), stavudine (d4T)/3TC, or tenofovir (TDF)/3TC in combination with either nevirapine (NVP) or efavirenz (EFV). Generalized Estimating Equation (GEE) model was used to identify risk factors associated with occurrence of ADR.

Results

2650 patients were followed-up for 2456 person-years and reported 114 ADRs (incidence rate = 4.6/100 person-years).There were more females 1706(64%) and 73(64%) of the ADRs were reported by women. Overall, 61(54%) of ADRs were reported by patients on AZT with 54(47%) of these occurring in patients on AZT/NVP. The commonest ADRs reported were pain 25(30%) and skinrash 10(18%). Most ADRs were grade 1(39%) with only 1% being life threatening (grade 4). Adjusted GEE analysis showed that ADR was less likely to occur in patients on longer duration of ART compared to the first six months on treatment; 6-12 months AOR 0.38(95% CI:0.16-0.91) and 12-24 months AOR 0.34(95% CI:0.16-0.73) respectively. Compared to patients on TDF, ADR was less likely to occur in patients on d4T and AZT AOR 0.18(95% CI 0.05-0.64) and AOR 0.24(95% CI:0.7-0.9) respectively. Age, gender and CD4 count were not significantly associated with ADRs.

Conclusion

ADRs are more likely to occur within the first six months on treatment. Close monitoring within this period is required to prevent occurrence of severe ADR and improve ART adherence. Further research on the tolerability of tenofovir in this environment is recommended.