Abstract: Blood-Brain-Barrier (BBB) is a strict permeability barrier for maintaining the Central Nervous System (CNS) homeostasis. One of the most important conditions to judge a CNS drug is to figure out whether it has BBB permeability or not. In the past 20 years, the existing prediction approaches are usually based on the data of the physical characteristics and chemical structure of drugs. However, these methods are usually only applicable to small molecule compounds based on passive diffusion through BBB. To deal this problem, one of the most famous methods is multi-core SVM method, which is based on clinical phenotypes about Drug Side Effects and Drug Indications to predict drug penetration of BBB. This paper proposed a Deep Learning method to predict the Blood-Brain-Barrier permeability based on the clinical phenotypes data. The validation result on three datasets proved that Deep Learning method achieves better performance than the other existing methods. The average accuracy of our method reaches 0.97, AUC reaches 0.98, and the F1 score is 0.92. The results proved that Deep Learning methods can significantly improve the prediction accuracy of drug BBB permeability and it can help researchers to reduce clinical trials and find new CNS drugs.

Abstract: Accurate prediction of drug-drug interactions (DDI) is a challenging task in drug discovery and development. It requires determination of enzyme inhibition in vitro which is highly system-dependent for many compounds. The aim of this study was to investigate whether the determination of intracellular unbound concentrations in primary human hepatocytes can be used to bridge discrepancies between results obtained using human liver microsomes and hepatocytes. Specifically, we investigated if Kpuu could reconcile differences in CYP enzyme inhibition values (Ki or IC50). Firstly, our methodology for determination of Kpuu was optimized for human hepatocytes, using four well-studied reference compounds. Secondly, the methodology was applied to a series of structurally related CYP2C9 inhibitors from a Roche discovery project. Lastly, the Kpuu values of three commonly used CYP3A4 inhibitors-ketoconazole, itraconazole, and posaconazole-were determined and compared to compound-specific hepatic enrichment factors obtained from physiologically based modeling of clinical DDI studies with these three compounds. Kpuu obtained in suspended human hepatocytes gave good predictions of system-dependent differences in vitro. The Kpuu was also in fair agreement with the compound-specific hepatic enrichment factors in DDI models and can therefore be used to improve estimations of enrichment factors in physiologically based pharmacokinetic modeling.

Abstract: The goal of Binding MOAD is to provide users with a data set focused on high-quality x-ray crystal structures that have been solved with biologically relevant ligands bound. Where available, experimental binding affinities (Ka, Kd, Ki, IC50) are provided from the primary literature of the crystal structure. The database has been updated regularly since 2005, and this most recent update has added nearly 7000 new structures (growth of 21%). MOAD currently contains 32,747 structures, composed of 9117 protein families and 16,044 unique ligands. The data are freely available on www.BindingMOAD.org. This paper outlines updates to the data in Binding MOAD as well as improvements made to both the website and its contents. The NGL viewer has been added to improve visualization of the ligands and protein structures. MarvinJS has been implemented, over the outdated MarvinView, to work with JChem for small molecule searching in the database. To add tools for predicting polypharmacology, we have added information about sequence, binding-site, and ligand similarity between entries in the database. A main premise behind polypharmacology is that similar binding sites will bind similar ligands. The large amount of protein-ligand information available in Binding MOAD allows us to compute pairwise ligand and binding-site similarities. Lists of similar ligands and similar binding sites have been added to allow users to identify potential polypharmacology pairs. To show the utility of the polypharmacology data, we detail a few examples from Binding MOAD of drug repurposing targets with their respective similarities.

Abstract: The use of zebrafish in whole organism phenotypic assays has become a valuable strategy throughout the drug discovery process. Zebrafish assays can be used not only to screen libraries of compounds at the earliest stages but also to evaluate advanced leads for their effects on specific biological pathways or for toxicity. However, when confronted with inactivity of a compound in a zebrafish assay, there are little data that can be used to judge if the compound is truly biologically inert or inactive due to a lack of permeability into the model organism. While medicinal chemistry principles suggest parameters that are predictive of human oral bioavailability, cellular permeability, and even bacterial permeability, there have been no such parameters developed for zebrafish absorption. To address this question, we compiled a set of 700 compounds reported in the literature to be active in zebrafish assays, evaluated their properties, and compared them to properties derived from a set of historical drugs and a set of recently approved oral drugs. While some properties overlap, the averages and 10th and 90th percentiles of molecular weight, octanol-water partition coefficient (logP), H-bond counts, and polar surface area for zebrafish-active compounds are statistically different from those of known drugs. This analysis should be useful to scientists interpreting structure-activity relationships based on data from zebrafish assays and help to inform the translation from fish to mammals.

Abstract: Microbes are nature's chemists, capable of producing and metabolizing a diverse array of compounds. In the human gut, microbial biochemistry can be beneficial, for example vitamin production and complex carbohydrate breakdown; or detrimental, such as the reactivation of an inactive drug metabolite leading to patient toxicity. Identifying clinically relevant microbiome metabolism requires linking microbial biochemistry and ecology with patient outcomes. Here we present MicrobeFDT, a resource which clusters chemically similar drug and food compounds and links these compounds to microbial enzymes and known toxicities. We demonstrate that compound structural similarity can serve as a proxy for toxicity, enzyme sharing, and coarse-grained functional similarity. MicrobeFDT allows users to flexibly interrogate microbial metabolism, compounds of interest, and toxicity profiles to generate novel hypotheses of microbe-diet-drug-phenotype interactions that influence patient outcomes. We validate one such hypothesis experimentally, using MicrobeFDT to reveal unrecognized gut microbiome metabolism of the ovarian cancer drug altretamine.

Abstract: BACKGROUND: Predicting drug-target interactions is time-consuming and expensive. It is important to present the accuracy of the calculation method. There are many algorithms to predict global interactions, some of which use drug-target networks for prediction (ie, a bipartite graph of bound drug pairs and targets known to interact). Although these algorithms can predict some drug-target interactions to some extent, there is little effect for some new drugs or targets that have no known interaction. RESULTS: Since the datasets are usually located at or near low-dimensional nonlinear manifolds, we propose an improved GRMF (graph regularized matrix factorization) method to learn these flow patterns in combination with the previous matrix-decomposition method. In addition, we use one of the pre-processing steps previously proposed to improve the accuracy of the prediction. CONCLUSIONS: Cross-validation is used to evaluate our method, and simulation experiments are used to predict new interactions. In most cases, our method is superior to other methods. Finally, some examples of new drugs and new targets are predicted by performing simulation experiments. And the improved GRMF method can better predict the remaining drug-target interactions.

Abstract: BACKGROUND: Cancer cells become immortalized through telomere maintenance mechanisms, such as telomerase reverse transcriptase (TERT) activation. In addition to maintaining telomere length, TERT activates manifold cell survival signaling pathways. However, telomerase-associated gene signatures in cancer remain elusive. METHODS: We performed a systematic analysis of TERT high (TERT(high)) and low (TERT(low)) cancers using multidimensional data from The Cancer Genome Atlas (TCGA). Multidimensional data were analyzed by propensity score matching weight algorithm. Coexpression networks were constructed by weight gene coexpression network analysis (WGCNA). Random forest classifiers were generated to identify cancer subtypes. RESULTS: The TERT(high)-specific mRNA expression signature is associated with cell cycle-related coexpression modules across cancer types. Experimental screening of hub genes in the cell cycle module suggested TPX2 and EXO1 as potential regulators of telomerase activity and cell survival. MiRNA analysis revealed that the TERT(high)-specific miR-17-92 cluster can target biological processes enriched in TERT(low) cancer and that its expression is negatively correlated with the tumor/normal telomere length ratio. Intriguingly, TERT(high) cancers tend to have mutations in extracellular matrix organization genes and amplify MAPK signaling. By mining the clinical actionable gene database, we uncovered a number of TERT(high)-specific somatic mutations, amplifications and high expression genes containing therapeutic targets. Finally, a random forest classifier integrating telomerase-associated multi-omics signatures identifies two cancer subtypes showed profound differences in telomerase activity and patient survival. CONCLUSIONS: In summary, our results depict a telomerase-associated molecular landscape in cancers and provide therapeutic opportunities for cancer treatment.

Abstract: BACKGROUND: Data on chronic kidney disease development in HIV infection is important towards building a comprehensive knowledge of HIV, ageing and polypharmacy in Africa. Several previous studies on tenofovir-associated kidney disease in Africa have shown conflicting results. This review summarises what is known about the development of kidney disease in HIV-positive African patients on tenofovir disoproxil fumarate (TDF)-containing ART. We set out to document the occurrence of kidney disease in HIV-positive Africans on TDF-containing ART in population-based studies and to evaluate the renal safety of TDF in Africans. METHODS: We conducted a systemic review using published studies which were identified through a computerized search of original research using the Medline/PubMed database, EMBASE, EBM Reviews, Proquest Google Scholar and Global Health reported from inception until 5 October 2017. Two reviewers independently abstracted the data and performed quality assessment of the included studies. We screened 595 articles and included 31 in the qualitative analysis performed. RESULTS: A total of 106 406 patients (of whom 66,681 were on Tenofovir) were involved in these 31 studies with sample sizes ranging from 30 to 62,230. Duration on tenofovir-containing ART ranged from those initiating ART at baseline to those who received TDF for up to 9 years. All but one of the studies involved only patients 16 years and older. The studies had differing definitions of kidney dysfunction and were of variable study design quality. The documented outcomes had substantial discrepancies across the studies, most likely due to methodological differences, study size and disparate outcome definitions. CONCLUSIONS: Our review identified studies in Africans reporting statistically significant renal function decline associated with TDF use but the clinical significance of this effect was not enough to contraindicate its continued use in ART regimens. Consistent with studies in other populations, patients are at greater risk if they have pre-existing renal disease and are more advanced in age. More research is needed on paediatric populations under 16 years of age. Trial registration This review was registered on Prospero (registration number CRD42018078717).

Abstract: SUMMARY: Computational drug target prediction has become an important process in drug discovery. Network-based approaches are commonly used in computational drug-target interaction (DTI) prediction. Existing network-based approaches are limited in capturing the contextual information on how diseases, drugs and genes are connected. Here, we proposed a context-sensitive network (CSN) model for DTI prediction by modeling contextual drug phenotypic relationships. We constructed a Drug-Side Effect Context-Sensitive Network (DSE-CSN) of 139 760 drug-side effect pairs, representing 1480 drugs and 5868 side effects. We also built a protein-protein interaction network (PPIN) of 15 267 gene nodes and 178 972 weighted edges. A heterogeneous network was built by connecting the DSE-CSN and the PPIN through 3684 known DTIs. For each drug on the DSE-CSN, its genetic targets were predicted and prioritized using a network-based ranking algorithm. Our approach was evaluated in both de novo and leave-one-out cross-validation analysis using known DTIs as the gold standard. We compared our DSE-CSN-based model to the traditional similarity-based network (SBN)-based prediction model. The results suggested that the DSE-CSN-based model was able to rank known DTIs highly. In a de novo cross-validation, the area under the receiver operating characteristic (ROC) curve was 0.95. In a leave-one-out cross-validation, the average rank was top 3.2% for known DTIs. When it was compared to the SBN-based model using the Precision-Recall curve, our CSN-based model achieved a higher mean average precision (MAP) (0.23 versus 0.19, P-value<1e-4) in a de novo cross-validation analysis. We further improved the CSN-based DTI prediction by differentially weighting the drug-side effect pairs on the network and showed a significant improvement of the MAP (0.29 versus 0.23, P-value<1e-4). We also showed that the CSN-based model consistently achieved better performances than the traditional SBN-based model across different drug classes. Moreover, we demonstrated that our novel DTI predictions can be supported by published literature. In summary, the CSN-based model, by modeling the context-specific inter-relationships among drugs and side effects, has a high potential in drug target prediction. AVAILABILITY AND IMPLEMENTATION: nlp/case/edu/public/data/DSE/CSN_DTI.

Abstract: In light of increased co-prescription of multiple drugs, the ability to discern and predict drug-drug interactions (DDI) has become crucial to guarantee the safety of patients undergoing treatment with multiple drugs. However, information on DDI profiles is incomplete and the experimental determination of DDIs is labor-intensive and time-consuming. Although previous studies have explored various feature spaces for in silico screening of interacting drug pairs, their use of conventional cross-validation prevents them from achieving generalizable performance on drug pairs where neither drug is seen during training. Here we demonstrate for the first time targets of adversely interacting drug pairs are significantly more likely to have synergistic genetic interactions than non-interacting drug pairs. Leveraging genetic interaction features and a novel training scheme, we construct a gradient boosting-based classifier that achieves robust DDI prediction even for drugs whose interaction profiles are completely unseen during training. We demonstrate that in addition to classification power-including the prediction of 432 novel DDIs-our genetic interaction approach offers interpretability by providing plausible mechanistic insights into the mode of action of DDIs.