Downloads backgroundDownloads background
Publications Search
Explore how scientists all over the world use DrugBank in their research.
Published on April 3, 2020
READ PUBLICATION →

Dense module searching for gene networks associated with multiple sclerosis.

Authors: Manuel AM, Dai Y, Freeman LA, Jia P, Zhao Z

Abstract: BACKGROUND: Multiple sclerosis (MS) is a complex disease in which the immune system attacks the central nervous system. The molecular mechanisms contributing to the etiology of MS remain poorly understood. Genome-wide association studies (GWAS) of MS have identified a small number of genetic loci significant at the genome level, but they are mainly non-coding variants. Network-assisted analysis may help better interpret the functional roles of the variants with association signals and potential translational medicine application. The Dense Module Searching of GWAS tool (dmGWAS version 2.4) developed in our team is applied to 2 MS GWAS datasets (GeneMSA and IMSGC GWAS) using the human protein interactome as the reference network. A dual evaluation strategy is used to generate results with reproducibility. RESULTS: Approximately 7500 significant network modules were identified for each independent GWAS dataset, and 20 significant modules were identified from the dual evaluation. The top modules included GRB2, HDAC1, JAK2, MAPK1, and STAT3 as central genes. Top module genes were enriched with functional terms such as "regulation of glial cell differentiation" (adjusted p-value = 2.58 x 10(- 3)), "T-cell costimulation" (adjusted p-value = 2.11 x 10(- 6)) and "virus receptor activity" (adjusted p-value = 1.67 x 10(- 3)). Interestingly, top gene networks included several MS FDA approved drug target genes HDAC1, IL2RA, KEAP1, and RELA, CONCLUSIONS: Our dmGWAS network analyses highlighted several genes (GRB2, HDAC1, IL2RA, JAK2, KEAP1, MAPK1, RELA and STAT3) in top modules that are promising to interpret GWAS signals and link to MS drug targets. The genes enriched with glial cell differentiation are important for understanding neurodegenerative processes in MS and for remyelination therapy investigation. Importantly, our identified genetic signals enriched in T cell costimulation and viral receptor activity supported the viral infection onset hypothesis for MS.
Published on April 2, 2020
READ PUBLICATION →

Quantification for non-targeted LC/MS screening without standard substances.

Authors: Liigand J, Wang T, Kellogg J, Smedsgaard J, Cech N, Kruve A

Abstract: Non-targeted and suspect analyses with liquid chromatography/electrospray/high-resolution mass spectrometry (LC/ESI/HRMS) are gaining importance as they enable identification of hundreds or even thousands of compounds in a single sample. Here, we present an approach to address the challenge to quantify compounds identified from LC/HRMS data without authentic standards. The approach uses random forest regression to predict the response of the compounds in ESI/HRMS with a mean error of 2.2 and 2.0 times for ESI positive and negative mode, respectively. We observe that the predicted responses can be transferred between different instruments via a regression approach. Furthermore, we applied the predicted responses to estimate the concentration of the compounds without the standard substances. The approach was validated by quantifying pesticides and mycotoxins in six different cereal samples. For applicability, the accuracy of the concentration prediction needs to be compatible with the effect (e.g. toxicology) predictions. We achieved the average quantification error of 5.4 times, which is well compatible with the accuracy of the toxicology predictions.
Published in March 2020
READ PUBLICATION →

Ecotoxicological study of six drugs in Aliivibrio fischeri, Daphnia magna and Raphidocelis subcapitata.

Authors: Lomba L, Lapena D, Ros N, Aso E, Cannavo M, Errazquin D, Giner B

Abstract: The presence of drugs in the environment is an emerging issue in the scientific community. It has been shown that these substances are active chemicals that consequently affect aquatic organisms and, finally, humans as end users. To evaluate the toxicity of these compounds and how they affect the environment, it is important to perform systematic ecotoxicological and physicochemical studies. The best way to address this problem is to conduct studies on different aquatic trophic levels. In this work, an ecotoxicological study of six drugs (anhydrous caffeine, diphenhydramine hydrochloride, gentamicin sulphate, lidocaine hydrochloride, tobramycin sulphate and enalapril maleate) that used three aquatic biological models (Raphidocelis subcapitata, Aliivibrio fischeri and Daphnia magna) was performed. Additionally, the concentration of chlorophyll in the algae R. subcapitata was measured. Furthermore, EC50 values were analysed using the Passino and Smith classification (PSC) method, which categorized the compounds as toxic or relatively toxic. All of the studied drugs showed clear concentration-dependent toxic effects. The toxicity of the chemicals depended on the biological model studied, with Raphidocelis subcapitata being the most sensitive species and Aliivibrio fischeri being the least sensitive. The results indicate that the most toxic compound, for all the studied biological models, was diphenhydramine hydrochloride. Graphical abstract.
Published on March 30, 2020
READ PUBLICATION →

The First 75 Days of Novel Coronavirus (SARS-CoV-2) Outbreak: Recent Advances, Prevention, and Treatment.

Authors: Yan Y, Shin WI, Pang YX, Meng Y, Lai J, You C, Zhao H, Lester E, Wu T, Pang CH

Abstract: The recent severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2, previously known as 2019-nCoV) outbreak has engulfed an unprepared world amidst a festive season. The zoonotic SARS-CoV-2, believed to have originated from infected bats, is the seventh member of enveloped RNA coronavirus. Specifically, the overall genome sequence of the SARS-CoV-2 is 96.2% identical to that of bat coronavirus termed BatCoV RaTG13. Although the current mortality rate of 2% is significantly lower than that of SARS (9.6%) and Middle East respiratory syndrome (MERS) (35%), SARS-CoV-2 is highly contagious and transmissible from human to human with an incubation period of up to 24 days. Some statistical studies have shown that, on average, one infected patient may lead to a subsequent 5.7 confirmed cases. Since the first reported case of coronavirus disease 2019 (COVID-19) caused by the SARS-CoV-2 on December 1, 2019, in Wuhan, China, there has been a total of 60,412 confirmed cases with 1370 fatalities reported in 25 different countries as of February 13, 2020. The outbreak has led to severe impacts on social health and the economy at various levels. This paper is a review of the significant, continuous global effort that was made to respond to the outbreak in the first 75 days. Although no vaccines have been discovered yet, a series of containment measures have been implemented by various governments, especially in China, in the effort to prevent further outbreak, whilst various medical treatment approaches have been used to successfully treat infected patients. On the basis of current studies, it would appear that the combined antiviral treatment has shown the highest success rate. This review aims to critically summarize the most recent advances in understanding the coronavirus, as well as the strategies in prevention and treatment.
Published on March 30, 2020
READ PUBLICATION →

A key genomic signature associated with lymphovascular invasion in head and neck squamous cell carcinoma.

Authors: Zhang J, Lin H, Jiang H, Jiang H, Xie T, Wang B, Huang X, Lin J, Xu A, Li R, Zhang J, Yuan Y

Abstract: BACKGROUND: Lymphovascular invasion (LOI), a key pathological feature of head and neck squamous cell carcinoma (HNSCC), is predictive of poor survival; however, the associated clinical characteristics and underlying molecular mechanisms remain largely unknown. METHODS: We performed weighted gene co-expression network analysis to construct gene co-expression networks and investigate the relationship between key modules and the LOI clinical phenotype. Functional enrichment and KEGG pathway analyses were performed with differentially expressed genes. A protein-protein interaction network was constructed using Cytoscape, and module analysis was performed using MCODE. Prognostic value, expression analysis, and survival analysis were conducted using hub genes; GEPIA and the Human Protein Atlas database were used to determine the mRNA and protein expression levels of hub genes, respectively. Multivariable Cox regression analysis was used to establish a prognostic risk formula and the areas under the receiver operating characteristic curve (AUCs) were used to evaluate prediction efficiency. Finally, potential small molecular agents that could target LOI were identified with DrugBank. RESULTS: Ten co-expression modules in two key modules (turquoise and pink) associated with LOI were identified. Functional enrichment and KEGG pathway analysis revealed that turquoise and pink modules played significant roles in HNSCC progression. Seven hub genes (CNFN, KIF18B, KIF23, PRC1, CCNA2, DEPDC1, and TTK) in the two modules were identified and validated by survival and expression analyses, and the following prognostic risk formula was established: [risk score = EXPDEPDC1 * 0.32636 + EXPCNFN * (- 0.07544)]. The low-risk group showed better overall survival than the high-risk group (P < 0.0001), and the AUCs for 1-, 3-, and 5-year overall survival were 0.582, 0.634, and 0.636, respectively. Eight small molecular agents, namely XL844, AT7519, AT9283, alvocidib, nelarabine, benzamidine, L-glutamine, and zinc, were identified as novel candidates for controlling LOI in HNSCC (P < 0.05). CONCLUSIONS: The two-mRNA signature (CNFN and DEPDC1) could serve as an independent biomarker to predict LOI risk and provide new insights into the mechanisms underlying LOI in HNSCC. In addition, the small molecular agents appear promising for LOI treatment.
Published on March 24, 2020
READ PUBLICATION →

Environmental Risk and Risk of Resistance Selection Due to Antimicrobials' Occurrence in Two Polish Wastewater Treatment Plants and Receiving Surface Water.

Authors: Giebultowicz J, Nalecz-Jawecki G, Harnisz M, Kucharski D, Korzeniewska E, Plaza G

Abstract: In this study, a screening of 26 selected antimicrobials using liquid chromatography coupled to a tandem mass spectrometry method in two Polish wastewater treatment plants and their receiving surface waters was provided. The highest average concentrations of metronidazole (7400 ng/L), ciprofloxacin (4300 ng/L), vancomycin (3200 ng/L), and sulfamethoxazole (3000 ng/L) were observed in influent of WWTP2. Ciprofloxacin and sulfamethoxazole were the most dominant antimicrobials in influent and effluent of both WWTPs. In the sludge samples the highest mean concentrations were found for ciprofloxacin (up to 28 mug/g) and norfloxacin (up to 5.3 mug/g). The removal efficiency of tested antimicrobials was found to be more than 50% for both WWTPs. However, the presence of antimicrobials influenced their concentrations in the receiving waters. The highest antimicrobial resistance risk was estimated in influent of WWTPs for azithromycin, ciprofloxacin, clarithromycin, metronidazole, and trimethoprim and in the sludge samples for the following antimicrobials: azithromycin, ciprofloxacin, clarithromycin, norfloxacin, trimethoprim, ofloxacin, and tetracycline. The high environmental risk for exposure to azithromycin, clarithromycin, and sulfamethoxazole to both cyanobacteria and eukaryotic species in effluents and/or receiving water was noted. Following the obtained results, we suggest extending the watch list of the Water Framework Directive for Union-wide monitoring with sulfamethoxazole.
Published on March 23, 2020
READ PUBLICATION →

How Sure Can We Be about ML Methods-Based Evaluation of Compound Activity: Incorporation of Information about Prediction Uncertainty Using Deep Learning Techniques.

Authors: Sieradzki I, Lesniak D, Podlewska S

Abstract: A great variety of computational approaches support drug design processes, helping in selection of new potentially active compounds, and optimization of their physicochemical and ADMET properties. Machine learning is a group of methods that are able to evaluate in relatively short time enormous amounts of data. However, the quality of machine-learning-based prediction depends on the data supplied for model training. In this study, we used deep neural networks for the task of compound activity prediction and developed dropout-based approaches for estimating prediction uncertainty. Several types of analyses were performed: the relationships between the prediction error, similarity to the training set, prediction uncertainty, number and standard deviation of activity values were examined. It was tested whether incorporation of information about prediction uncertainty influences compounds ranking based on predicted activity and prediction uncertainty was used to search for the potential errors in the ChEMBL database. The obtained outcome indicates that incorporation of information about uncertainty of compound activity prediction can be of great help during virtual screening experiments.
Published on March 23, 2020
READ PUBLICATION →

Analysis of Dual Class I Histone Deacetylase and Lysine Demethylase Inhibitor Domatinostat (4SC-202) on Growth and Cellular and Genomic Landscape of Atypical Teratoid/Rhabdoid.

Authors: Hoffman MM, Zylla JS, Bhattacharya S, Calar K, Hartman TW, Bhardwaj RD, Miskimins WK, de la Puente P, Gnimpieba EZ, Messerli SM

Abstract: Central nervous system atypical teratoid/rhabdoid tumors (ATRTs) are rare and aggressive tumors with a very poor prognosis. Current treatments for ATRT include resection of the tumor, followed by systemic chemotherapy and radiation therapy, which have toxic side effects for young children. Gene expression analyses of human ATRTs and normal brain samples indicate that ATRTs have aberrant expression of epigenetic markers including class I histone deacetylases (HDAC's) and lysine demethylase (LSD1). Here, we investigate the effect of a small molecule epigenetic modulator known as Domatinostat (4SC-202), which inhibits both class I HDAC's and Lysine Demethylase (LSD1), on ATRT cell survival and single cell heterogeneity. Our findings suggest that 4SC-202 is both cytotoxic and cytostatic to ATRT in 2D and 3D scaffold cell culture models and may target cancer stem cells. Single-cell RNA sequencing data from ATRT-06 spheroids treated with 4SC-202 have a reduced population of cells overexpressing stem cell-related genes, including SOX2. Flow cytometry and immunofluorescence on 3D ATRT-06 scaffold models support these results suggesting that 4SC-202 reduces expression of cancer stem cell markers SOX2, CD133, and FOXM1. Drug-induced changes to the systems biology landscape are also explored by multi-omics enrichment analyses. In summary, our data indicate that 4SC-202 has both cytotoxic and cytostatic effects on ATRT, targets specific cell sub-populations, including those with cancer stem-like features, and is an important potential cancer therapeutic to be investigated in vivo.
Published on March 18, 2020
READ PUBLICATION →

SAEROF: an ensemble approach for large-scale drug-disease association prediction by incorporating rotation forest and sparse autoencoder deep neural network.

Authors: Jiang HJ, Huang YA, You ZH

Abstract: Drug-disease association is an important piece of information which participates in all stages of drug repositioning. Although the number of drug-disease associations identified by high-throughput technologies is increasing, the experimental methods are time consuming and expensive. As supplement to them, many computational methods have been developed for an accurate in silico prediction for new drug-disease associations. In this work, we present a novel computational model combining sparse auto-encoder and rotation forest (SAEROF) to predict drug-disease association. Gaussian interaction profile kernel similarity, drug structure similarity and disease semantic similarity were extracted for exploring the association among drugs and diseases. On this basis, a rotation forest classifier based on sparse auto-encoder is proposed to predict the association between drugs and diseases. In order to evaluate the performance of the proposed model, we used it to implement 10-fold cross validation on two golden standard datasets, Fdataset and Cdataset. As a result, the proposed model achieved AUCs (Area Under the ROC Curve) of Fdataset and Cdataset are 0.9092 and 0.9323, respectively. For performance evaluation, we compared SAEROF with the state-of-the-art support vector machine (SVM) classifier and some existing computational models. Three human diseases (Obesity, Stomach Neoplasms and Lung Neoplasms) were explored in case studies. As a result, more than half of the top 20 drugs predicted were successfully confirmed by the Comparative Toxicogenomics Database(CTD database). This model is a feasible and effective method to predict drug-disease correlation, and its performance is significantly improved compared with existing methods.
Published on March 18, 2020
READ PUBLICATION →

A deep learning-based method for drug-target interaction prediction based on long short-term memory neural network.

Authors: Wang YB, You ZH, Yang S, Yi HC, Chen ZH, Zheng K

Abstract: BACKGROUND: The key to modern drug discovery is to find, identify and prepare drug molecular targets. However, due to the influence of throughput, precision and cost, traditional experimental methods are difficult to be widely used to infer these potential Drug-Target Interactions (DTIs). Therefore, it is urgent to develop effective computational methods to validate the interaction between drugs and target. METHODS: We developed a deep learning-based model for DTIs prediction. The proteins evolutionary features are extracted via Position Specific Scoring Matrix (PSSM) and Legendre Moment (LM) and associated with drugs molecular substructure fingerprints to form feature vectors of drug-target pairs. Then we utilized the Sparse Principal Component Analysis (SPCA) to compress the features of drugs and proteins into a uniform vector space. Lastly, the deep long short-term memory (DeepLSTM) was constructed for carrying out prediction. RESULTS: A significant improvement in DTIs prediction performance can be observed on experimental results, with AUC of 0.9951, 0.9705, 0.9951, 0.9206, respectively, on four classes important drug-target datasets. Further experiments preliminary proves that the proposed characterization scheme has great advantage on feature expression and recognition. We also have shown that the proposed method can work well with small dataset. CONCLUSION: The results demonstration that the proposed approach has a great advantage over state-of-the-art drug-target predictor. To the best of our knowledge, this study first tests the potential of deep learning method with memory and Turing completeness in DTIs prediction.