2011

2011

December

  • Translating Clinical Findings into Knowledge in Drug Safety Evaluation – Drug Induced Liver Injury Prediction System (DILIps). Liu Z. et al. Plos Comput Biol. (5448)
  • KEYWORDS: ASSAY DATA – DATABASES – DILI – DRUG SAFETY – HEPATOTOXICITY – QSAR MODELING – TEXT-MINING

    A new predictive system devoted to drug induced liver injury prediction (DILIps) is presented as a recent example of challenges to bridge preclinical and clinical findings in the case of hepatotoxicity data (extracted from SIDER database (888 drugs), DrugBank database (6620 drugs), and three literature datasets: LTKB benchmark set (287 drugs), Pfizer hepatotoxicity set (626 compounds), and O’Brien set (135 drugs); see Support Material table S6). Based on 13 hepatotoxic side effects identified (bilirubinemia, cholecystitis, cholelithiasis, cirrhosis, elevated liver function tests, hepatic failure, hepatic necrosis, hepatitis, hepatomegaly, jaundice, liver disease, liver fatty, liver function tests abnormal) and the ‘Rule of three’ (consider drugs data when they are incriminated by 3 or more hepatotoxic side effects defined), the respective models were built in order to distinguish prediction between different findings described (see Table 1). The list of PMIDs as literature proof about the co-ocurrence between the 13 hepatotox side effects and protein targets are provided as result of text-mining approaches application.

  • Evaluation of Drugs with Specific Organ Toxicities in Organ Specific Cell Lines. Lin Z. et al. Toxicol Sci. (5443)
    Involved Partner: PFIZER
  • KEYWORDS: ASSAY DATA – CARDIOTOXICITY – HEPATOTOXICITY – NEPHROTOXICITY

    Based on data for 273 hepatotoxic, 191cardiotoxic, and 85 nephrotoxic compounds in HepG2 (hepatocellular carcinoma), H9c2 (embryonic myocardium), and NRK-52E (kidney proximal tubule) cells for evaluation of their cytotoxicity, the results of this study show how select organ toxicity potentially results form compound accumulation in a particular tissue, cell types within organs, metabolism and off-target effects, and highlights that consideration of Cmax values improves significantly the prediction of toxicities.

  • BOOK CHAPTER: Drug-Induced Phospholipidosis. Bernstein PR. et al. Annu Rep Med Chem. (5441)
  • KEYWORDS: ASSAY DATA – PHOSPHOLIPIDOSIS

    This book chapter reviews the current knowledge to explain the mechanism of the phospholipidosis event and the different techniques, both in silico (ROCHE: CAFCA, Organon: based on clogP and pKa, Pfizer and FDA performed QSAR models based on MCAPC-QSAR and MDL-QSAR) and in vitro (see section 3.3), available to predict this type of drug induced toxicity. Examples of pharmacological profiling projects are detailed (ROCHE evaluated DPP-IV inhibitors set, AstraZeneca a series of 5-HT1B antagonists, ROCHE and JNJ carried out analysis of H3 antagonists) as related activity research in pharmaceutical companies to anticipate the phospholipidosis occurrence.

  • Analysis of Commercial and Public Bioactivity Databases. Tiikkainen T. et al. J Chem Inf Model. (5440)
  • KEYWORDS: DATABASES – ENZYMES – NUCLEAR RECEPTORS – PHARMACOLOGY – TRANSPORTERS

    Description of the Metadabase resulting from the integration of different public (ChEMBL, PubChem and PDSP Ki) and commercial (WOMBAT and Evolvus) databases, developed as a MySQL relational database with different entities (molecules, targets and activities) where the InChI key calculation helps to avoid redundancy of structures and to cluster related information. See Table 1 for general statistics of the Metadabase and Table 3 for a distribution of activities across five major protein classes (enzymes, GPCRs, ion channels, nuclear receptors, and transporters).

  • Drug discovery in the age of systems biology: the rise of computational approaches for data integration. Iskar M. et al. Curr Opin Biotechnol. (5438)
  • KEYWORDS: DATABASES – DRUG DISCOVERY – PATHWAYS – SYSTEMS BIOLOGY

    To advance in the understanding of drugs action a proposal of integration of data from different public resources (PubChem, ChEMBL, DrugBank, SITITCH, GEO, Connectivity Map, ACTor, SIDER and NPC, see Table 1 for links and description of contents) is emphasized in this article to encourage current initiatives to interlink data from chemical-protein interactions, binding affinities, mode of action, transcriptomic/proteomic/metabolic/side_effects profiles, ADME/toxicology and patient records data.

  • In-silico assay for assessing phospholipidosis potential of small drug like molecules: training, validation and refinement using several datasets. Fischer H. et al. J Med Chem. (5437)
  • KEYWORDS: ASSAY DATA – PHOSPHOLIPIDOSIS

    A new study to predict phospholipidosis potential, based on measurements of the Free Energy of Amphiphilicity (see tables 3 and 4), in silico> pKa calculation or data from phospholipidosis in vitro assay applied to a set of 32 drugs (see Table 2 for compounds names). The new models were validated based on assessment of 422 compounds via in-house in vitro and in silico methods, a FDA of 91 compounds, and with specific physicochemical properties (lipophilicity, see Table 1)).

November

  • The value of in silico chemistry in the safety assessment of chemicals in the consumer goods and pharmaceutical Industries. Modi S. et al. Drug Discov Today. (5423) eTOX cited
  • KEYWORDS: DATABASES – DRUG SAFETY – QSAR MODELING – SOFTWARE – STRUCTURE-BASED PREDICTION – RISK ASSESSMENT

    Discussion on limitations and strengths of in silico tools, where examples of public data sources as useful training sets for predictive chemistry models (see Table 1, ACToR, CCRIS, ChEMBL, CTD, DART, NTP, RepDOSE and ToxRefDB) are presented to discuss the choosing of the right data, and a list of free open source tools/software for predictive toxicity models (see Table 2, CAESAR, Lazar, OECD Toolbox, OncoLogic, Tox-Comp, ToxTree and VirtualToxLab) is provided together with discussion on the choice of methos, the importance of applicability domain and the limitations of QSAR approaches.

October

  • Development of an in vitro liver toxicity prediction model based on longer term primary rat hepatocyte culture. Hrach J. et al. Toxicol Lett. (5421)
  • KEYWORDS: ASSAY DATA – ENZYMES – HEPATOTOXICITY – TOXICOGENOMICS

    According to the genomic profile of compounds, authors shows how combination of hepatocyte sandwich culture and global gene expression analysis can help to define a model for hepatotoxicity prediction for unknown compounds. This study provides the top ‘hit’ genes with some role in hepatotoxicity events and their known function (see Table 2: Proteasome complex, cyclin-dependent kinase (Cdk7) or s-phase related proteins, Myc, Egf, Map kinase activated protein kinase 2 (Mapkapk2), Tgfβ2 and kappaB kinase, inhibitor (Ikbkb), Lactate dehydrogenase B (Lahb), triosephosphate isomerase (Tim) and enolase, ATP synthase C1 and subunits, cytochrome c reductases NADH dehydrogenases, Junctiona adhesion molecules 3 (Jan3) and claudin 10, CYP1A1, CYP2A1 and Microsomal Gst 2).

  • In vitro-in vivo extrapolation of clearance: Modeling hepatic metabolic clearance of highly bound drugs and comparative assessment with existing calculation methods. Poulin P. et al. J Pharm Sci. (5420)
  • KEYWORDS: ASSAY DATA – PHARMACOKINETICS

    Data compilation corresponding to a human dataset of 25 drugs for the prediction of Clearance (see Table 1 for logP, pKa, F1 (ratio of fraction unionized between plasma (pH=7.4) and intracellular water (pH=7.0), …, main binding proteins, scaled microsomal clearance and human plasma clearance in vivo values). This study demonstrates the influence of drug ionization and differential binding of highly bound drugs when located in the liver.

  • Novel In Vitro-In Vivo Extrapolation (IVIVE) Method to Predict Hepatic Organ Clearance in Rat. Umehara KI. et al. Pharm Res. (5419)
    Involved Partner: NVS
  • KEYWORDS: ASSAY DATA – PHARMACOKINETICS

    Report of data (Kinetic Parameters for the Uptake into Suspended Hepatocytes (Table I), kinetic parameters for the metabolism in liver microsomes (Table II), kinetic parameters for the biliary excretion in sandwich-cultured hepatocytes (Table III), kinetic parameters for the active sinusoidal efflux from hepatocytes (Table IV) an up-scaled and predicted hepatic clearances from in vitro assays, in vitro and in vivo pharmacokinetic parameters (Table VI)) for a set of 13 chemicals (propranolol, quinidine, verapamil, cyclosporine A, ketoconazole, atorvastatin, aliskiren, pravastatin, valsartan, benzylpenicillin, digoxin, furosemide and ciprofloxacin).

  • Comparing bioassay response and similarity ensemble approaches to probing protein pharmacology. Chen B. et al. Bioinformatics. (5418)
    Involved Partner: PFIZER
  • KEYWORDS: PHARMACOLOGY

    Based on a set of 5672 unique compounds from 155 Pfizer internal BioPrint assays, authors present comparison of results between a creation of a protein network based on ligand similarity approach application and ligand bioassay response-data as reference data. Both approaches provided well-known protein-protein interactions and also new relationships. See the example of 5-HT2B results in Table 3.

  • Analysis of altered gene expression specific to embryotoxic chemical treatment during embryonic stem cell differentiation into myocardiac and neural cells. Suzuki N. et al. J Toxicol Sci. (5416)
  • KEYWORDS: ASSAY DATA – REPRODUCTIVE AND DEVELOPMENTAL TOXICITY

    Example of data from an altered gene expression evaluation during a Embryonic stem cells differentiation into myocardiac and neural cells on treatment with some embryotoxic and non-embryotoxic chemicals (see Table for a list of 12 selected test chemicals, and tables 2-5 for assay data). This study reports genes upregulated that have an indispensable role for differentiation and development of heart and brain tissues that can explain possible embryotoxicity events at least at in vitro level.

  • Cytochrome P450 enzymes in the brain: emerging evidence of biological significance. Ferguson CS. et al. Trends Pharmacol Sci. (5415)
  • KEYWORDS: CYP450 – ENZYMES

    This article provides a briefly overview of current knowledge to develop new strategies for understanding of the roles and regulation of brain to better predict, prevent and treat disease. The authors summarize, with some examples, the CYPs expressed in the brain grouped according to their centrally acting substrates (clinical drugs, neurotoxins, drug abuse, fatty acids, steroids and neurotransmitters; see Table 1 and section ‘Insight into the functional role of CYPs in the brain’).

  • Gaussian interaction profile kernels for predicting drug-target interaction. van Laarhoven T. et al. Bioinformatics. (5414)
  • KEYWORDS: DATABASES – PHARMACOLOGY – STRUCTURE-BASED PREDICTION

    A new method (Gaussian Interaction Profile (GIP) kernel combined with a simple classifier, (kernel) Regularized Least Squares (RLS)) to assess drug-protein interactions, based on data from KEGG BRITE, BRENDA, SuperTarget and DrugBank databases. Methodology details are widely provided in section 2, 3 and 4; and examples of results (see tables 3 and 4).

  • Empowering industrial research with shared biomedical vocabularies. Harland L. et al. Drug Discov Today. (5413)
    Involved Partner: GSK, AZ, ROCHE, PFIZER
  • KEYWORDS: DATA MINING – TEXT MINING

    his article reviews emerging needs and challenges to avoid replication of efforts to construct and maintain vocabularies along industries and academia institutions. In order to establish basic knowledge for these attempts of data harmonization, a comprehensive description of key concepts is provided in Box 1 (definition of vocabulary, basic identity (domain, identifier, preferred name, definition), organizational structure (dictionaries, thesauri, taxonomies, ontology), linguistic features (synonyms, homonyms, antonyms, ‘part of speech’ fragments) and other important elements (provenance, cross-references, and domain-specific features)). Authors gather in Table 2 a list of potential nonprofit partners for industry vocabulary management.

September

  • What makes a good drug target?. Gashaw I. et al. Drug Discov Today. (5409)
  • KEYWORDS: PHARMACOLOGY

    Based on a classification of targets and their modes of action (see Table 1) and Bayer HealthCare experience, this review reports the pillars of the target evaluation process (see Figure 1, Target assessment: Target identification, Target validation, Druggability assessment, and Assayability assessment), a list of Target identification strategies (see Figure 2) as different aspects to be evaluated in order to define a good drug target. In addition, they comment on the clinical and commercial needs and the intellectual property situation.

  • Challenges for computational structure-activity modelling for predicting chemical toxicity: future improvements?. Combes RD. Expert Opin Drug Metab Toxicol. (5408)
  • KEYWORDS: MOLECULAR DESCRIPTORS – STRUCTURE-BASED PREDICTION

    As a cost-effective approach for predicting chemical toxicity, the computational toxicology methods attempt also to help in prioritising chemicals for testing advance. Different problems in the development of computational models are covered in this review: i) the use of inappropriate molecular descriptors and tools that are not transparent (see section 2); ii) the undetected existence of chemicals that cause large changes in toxicity with only small differences in molecular structure (causing ‘activity cliffs’ in the structure-activity landscape) (see section 3); iii) spurious correlations between structure and activity (see section 4); iv) lack of quality control of toxicity data (see section 5); v) difficulties in determining predictivity for novel chemicals; and vi) an over-reliance on complex mathematics and statistics.

  • Collation and data-mining of literature bioactivity data for drug discovery. Bellis LJ. et al. Biochem Soc Trans. (5406)
    Involved Partner: EMBL
  • KEYWORDS: DATABASES – DATA MINING – PHARMACOLOGY – PHARMACOKINETICS

    A detailed overview of the collating process to populate the current release of ChEMBL database (955.004 dinstinct compounds, 8.370 targets) is reported, together with explanations of how all this data can be searchable (using an interactive compound sketch tool, SMILES strings, gene identifiers, protein sequence similarity, protein families, etc) in a single repository. The whole database contains 3 part databases: ChEMBL-NTD with open data primary screening and medicinal chemistry data for neglected diseases; Kinase SARfari with an integrated chemogenomics workbench focused on protein kinases that contains data donated by Novartis, GlaxoSmithKline and St Jude Children’ Research Hospital; and GPCR SARfari with an integrated chemogenomics workbench focused on rhodopsin-like GPCRs.

  • ChEMBL: a large-scale bioactivity database for drug discovery. Galton A. et al. Nucl. Acids Res. (5405)
    Involved Partner: EMBL
  • KEYWORDS: DATABASES – PHARMACOLOGY

    An update of open data database ChEMBL which currently contains information of binding,
    functional and ADMET information for a large number of drug-like bioactive compounds (5.4 million bioactivity measurementsfor more than 1 million compounds and 5200 protein targets, see Table 1 for a list of sources of compound and bioactivity data in ChEMBL_11). Authors describe details regarding the data extraction (from a variety of journals, such as Journal of Medicinal Chemistry, Bioorganic Medicinal Chemistry Letters and Journal of Natural Products) and curation; highlight the inclusion of structures and annotation for FDA-approved drugs, and extend explanations about download and web services functionalites to facilitate query the database locally.

  • Integrating GPCR-specific information with full text articles. Vroling B. et al. BMC Bioinformatics. (5404)
  • KEYWORDS: DATABASES – DATA MINING – PHARMACOLOGY – TEXT MINING – SOFTWARE

    In order to not miss any data related with GPCR, this article presents the efforts to identify relevant information (e.g., protein identification and normalization, or mutation and residue identification) disseminated in a certain article to extract more knowledge and have useful links to other resources devoted to GPCR data.

  • Strategies to improve in vivo toxicology outcomes for basic candidate drug molecules. Luker T. et al. Bioorg Med Chem Lett. (5403)
    Involved Partner: AZ
  • KEYWORDS: ASSAY DATA – STRUCTURE-BASED PREDICTION

    Based on a set of marketed post-2000 basic drugs, AstraZeneca shows their success to classify basic candidate drug molecules to predict attrition in pre-clinical toxicology stage (see Table 2 for predicted outcomes of 25 marketed drug bases) using a valid PLS-DA considering the list of calculated descriptors provided in Table 1 where definitions are detailed.

  • Pharmacogenetics: past, present and future. Pirmohamed M. Drug Discov Today. (5401)
  • KEYWORDS: DRUG DISCOVERY

    This article offers an historical overview of important advances regarding the identification of genetic factors to understand the drug response (see Table 1), provides the most significant genetic predictors of drug response (see Table 2) and highlights the possible reasons to argue the difficulties in the translation of pharmacogenetic findings into the clinical practice (see Box 1). Interestingly, authors reports in a single picture the many different –omics terms to locate the future of the pharmacogenomics.

August

  • Evaluation of Three State-of-the-art Metabolite Prediction Software Packages (Meteor, MetaSite, and StarDrop) through Independent and Synergistic Use. Tjollyn H. et al. Drug Metab Dispos. (5398)
    Involved Partner: J&J
  • KEYWORDS: ENZYMES – METABOLISM – SOFTWARE

    Interesting article to have a comprehensive comparison (see Table 1) between 3 software packages developed for metabolite prediction (Meteor (rule-based tool), MetaSite (automated docking model with reactivity correction designed to predict Phase I CYP450 metabolism) and StarDrop (based on quantum mechanical approach for the prediction of the relative involvement of CYP3A4, CYP2D6 and CYP2C9)). Based on a test set of 22 compounds (see Table 2 for compound name, indication/target, single oral dose, metabolic clearance normalized and type of biotransformation, and Figure 2 for compound structures) authors provide comparison of results for each combination of software pieces (intersection or union), see Table 3.

  • Potential carcinogenic hazards of non-regulated disinfection by-products: Haloquinones, halo-cyclopentene and cyclohexene derivatives, N-halamines, halonitriles, and heterocyclic amines. Bull RJ. et al. Toxicology. (5397)
  • KEYWORDS: ASSAY DATA – CARCINOGENICITY – STRUCTURE-BASED PREDICTION

    This review reports data (see tables 2, 3, 4, 5 and 6) regarding different sets of disinfection by-products (DBPs) and their carcinogenic potential since there is the suspicion of chlorination of drinking water can derive in bladder cancer occurrence. Their results do not clearly support this thesis, but authors shed the light that reasons can be explained by their corresponding metabolites and suggest further investigations.

  • Identifying Compound-Target Associations by Combining Bioactivity Profile Similarity Search and Public Databases Mining. Cheng T. et al. J Chem Inf Model. (5396)
  • KEYWORDS: DATABASES – DATA MINING – MOLECULAR DESCRIPTORS – PHARMACOLOGY – STRUCTURE-BASED PREDICTION

    A new computational approach, bioactivity profile similarity search (BASS), is presented to facilitate identification of neighbor compounds based on their similar bioactivity profile to define new compound-target associations. Consult of different public chemical annotated libraries (DrugBank, TTD, ChEMBL and PubChem) let authors to conduct their approach for a set of 4296 compounds (see Table 2 of Support Material) that form their bioactivity profile database. Interestingly, the results evidence also the power of suggesting novel chemical scaffolds for specific targets that this approach could mean in the field of the drug discovery.

  • Literature analysis for systematic drug repurposing: a case study from Biovista. Lekka E. et al. Drug Discov Today. (5395)
  • KEYWORDS: DATABASES – DRUG DISCOVERY – SOFTWARE – TEXT MINING

    To exemplify the potential of the literature-based discovery strategies, a case study focused on the treatment of Multiple Sclerosis is presented as application of the Biovista software, which relies on an extensive proprietary database of context-crossing relations among biomedical entities as diseases, ADRs, drugs, compounds, genes, pathways, among others, from Genomic databases (EntrezGene, UniProt, PDB and Gene Expression Atlas); Microarray repositories (GEO and ArrayExpress); Pathway databases KEGG Pathway, Reactome, (NCI)-Nature Pathway Interaction Database); Cheminformatics resources (PubChem, ChEMBL and ChemSpider); Drug-to-target databases (DrugBank, PharmGKB and MATADOR); Drug-related-to-disease databases (Drugs@FDA database, AERS, SIDER and DailyMed).

  • Cheminformatic/bioinformatic analysis of large corporate databases: Application to drug repurposing. Loging W. et al. Drug Discov Today. (5393)
  • KEYWORDS: DRUG DISCOVERY – DATABASES – PHARMACOLOGY

    This article reviews the high-level strategies for drug repositioning (in vivo and in vitro screening, gene expression screening, systemic expert review, in silico discovery (data integration and network based)) with examples of different mining approaches and pros and cons aspects (see Table 1A), and provides a list of potential public source data available for their application (Pubmed, Online Mendelian Inheritance in Man, Mouse Genome Informatics, Kyoto Encyclopedia of Genes and Genome, BioCarta, IUPHAR database, ChEMBL, PubChem, ClinicalTrials.gov and SNOMED-CT; see Table 1B).

  • In silico prediction of rhabdomyolysis of compounds by self-organizing map and support vector machine. Hu X. et al. Toxicol In Vitro. (5392)
  • KEYWORDS: ASSAY DATA – MOLECULAR DESCRIPTORS – STRUCTURE-BASED PREDICTION

    Based on a set of 303 drugs and toxins (186 positive and 117 negative), authors develop classification models to predict rhabdomyolysis, a musculoskeletal toxicity event, by application of Kohonen’s self-organizing map and support vector machine methodologies (see Support Material table with data associated) using the program ADRIANAcode to to calculate molecular descriptors and proceed the study with a selection of 16 due to their better correlation with the activity.

  • The role of fragment-based and computational methods in polypharmacology. Bottegoni G. et al. Drug Discov Today. (5391)
  • KEYWORDS: DRUG DISCOVERY – PHARMACOLOGY – STRUCTURE-BASED PREDICTION

    This article reviews several computational approaches to facilitate optimally the advance in different ways to address the new drug discovery paradigm, the multitarget strategy. The authors propose a fragment-based pipeline for multitarget discovery, from a virtual library to in vivo models and pharmacokinetic profiling through multitarget fragment recognition to multitarget lead proposal (see Figure 4).

  • Systems genetics for drug target discovery. Penrod NM. et al.Trends Pharmacol Sci. (5390)
  • KEYWORDS: NETWORKS – PHARMACOLOGY – SYSTEMS BIOLOGY

    Current insights of networks exploitation in the field of systems biology focused in genetic data, a descriptive review of 1) genome-wide association studies, mapping the genetic basis of diseases; 2) validation of functional consequences of genetic variation, form disease-associated genomic regions to functional targets (functional genetics and genomics); 3) genome-wide regulatory networks, as bridge of gap between disease-associated variants and disease-associated genes; and 4) pharmacology networks, interlink between disease-associated genes and functional targets.

  • Novel advances in cytochrome P450 research. Singh D. et al. Drug Discov Today. (5389)
  • KEYWORDS: CYP450

    A comprehensive update on CYP450 research about their role in drug metabolism, their expression regulation by non-coding RNA, some examples of in silico miRNA predictions indicating their post-transcriptional regulation, the miRNAs and nuclear receptors that regulates specific CYPs activity (see Table 1), and recent developments to conduct CYP genotype-guide therapies (e.g., warfarin and taximofen)..

  • A Data Mining Method To Facilitate SAR Transfer. Bajorath J. et al. J Chem Inf Model. (5387)
  • KEYWORDS: DATA MINING – DATABASES – DRUG DISCOVERY – QSAR PREDICTION – STRUCTURE-BASED PREDICTION

    This study presents a new text-mining approach to explore alternative chemotypes that display similar SAR characteristics and potency progression in a chemical collection. The results obtained to identify chemical series with SAR transfer potential if one analog series is available as a starting point and to detect all SAR transfer events that occur in compound datasets are reported based on BindingDB dataset contents. Four SAR transfer series are discussed in more detail (Dopamine D3 receptor antagonists (Figure 3), Thrombin inhibitors (Figure 4), Factor Xa inhibitors (Figure 5) and Carbonic Anhydrase I inhibitors (Figure 6).

  • Characterization of steroid hormone receptor activities in 100 hydroxylated polychlorinated biphenyls, including congeners identified in humans. Takeuch S. et al. Toxicology. (5386)
  • KEYWORDS: ASSAY DATA – NUCLEAR RECEPTORS – PHARMACOLOGY

    This article reports data of hormonal activity of a set of 100 hydroxylated polychlorinated biphenyls against 4 nuclear receptors (estrogen alpha and beta, androgen, and glucocorticoid receptors). Name and values for each compound about their agonist and antagonist mode of action activity and their structural classification are provided in tables 1 and 2.

  • Real external predictivity of QSAR models: how to evaluate it? A comparison of different validation criteria and proposal of using the concordance correlation coefficient. Chirico N. et al. J Chem Inf Model. (5384)
  • KEYWORDS: QSAR MODELLING

    A new approach for the external validation of a QSAR model is presented in this article, it is based on a simple statistical parameter, the concordance correlation coefficient, used as a criteria for a single validation measure. Comparison with current state-of-the-art methodologies results showed a good agreement of its performance after analyzing 210,000 data sets.

  • Structure-based Site of Metabolism Prediction for Cytochrome P450 2D6. Moors SL. et al. J Med Chem. (5382)
  • KEYWORDS: CYP450 – METABOLISM – STRUCTURE-BASED PREDICTION

    A combination of structure-based methodologies (protein ensemble generation and ensemble docking) to provide prediction of the metabolism activity in the case of a 54 CYP2D6 substrates set.

  • Predicting Phospholipidosis: A fluorescence non-cell based in vitro assay for the determination of drug-phospholipid complex formation in early drug discovery. Zhou L. et al. Anal Chem. (5381)
  • KEYWORDS: ASSAY DATA – PHOSPHOLIPIDOSIS

    This article reports the validation exercise of using a high-fluorescence non-cell based assay to evaluate the drug-phospholipid interaction potential of a given compound. Comparison with the current known phospholipidosis induced agents and their respective metabolites (see tables 1 and 2) evidences a good correlation to predict phospholipidosis occurrence.

  • A Quality Alert and Call for Improved Curation of Public Chemistry Databases. Williams AJ. et al. Drug Discov Today. (5380)
  • KEYWORDS: DATABASES – PHARMACOLOGY

    A critical report to highlight the questionable quality of the diverse internet-based chemistry resources, the authors recommend to contrast data between different sources (PubChem, ChemIDPlus, EPA’s ACToR, ChEMBL, DrugBank, the Human Metabolome Database, KEGG, ChemSpider, etc.).

  • Evaluation and validation of multiple cell lines and primary mouse macrophages to predict phospholipidosis potential. Lecureux L. et al. Toxicol In Vitro. (5379)
  • KEYWORDS: ASSAY DATA – PHOPHOLIPIDOSIS – RISK ASSESSMENT

    Data of in vitro phospholipidosis activity in different cells lines (HepG2, HuH7, I-1335 and RAW264.7) is provided for a set of 30 known drugs and 21 structure unknown compounds (see Table 2) and results analysis points out that for a progression of a compound that causes phospholipidosis in vivo several aspects should be taken into account to judge risk assessmen, if there is intended clinical indication, safety multiples, its severity, sequaele to toxicity, multiplicity of target organ effected, and /or even types of target organs affected to assess this kind of toxicity, weather phospholipidosis is still under debate if it is a toxic or adaptive response.

  • Integrated decision support for assessing chemical liabilities. Spjuth O. et al. J Chem Inf Model. (5378)
    Involved Partner: AZ
  • KEYWORDS: ADME – ASSAY DATA – RISK ASSESSMENT – SOFTWARE

    Presentation of architecture (see Figure 1) and application of the Bioclipse Decision Support system, developed by AstraZeneca, in three important safety endpoints (mutagenicity (data from an Ames test: 4337 structures (2401 mutagen)), carcinogenicity (data from CPDB: 892 structures (470 actives)) and aryl hydrocarbon receptor activation (data from PubChem BioAssay: 15,951 structures (7971 actives))) shows the potential of such tool, available online at http://www.bioclipse.net/decision-support.

July

  • Understanding pharmacology in humans: Phase I and Phase II (Data Generation). Merlo Pich E. Curr Opinion in Pharmacol. (5376)
  • KEYWORDS: DRUG DISCOVERY

    This article reports the advantages of the integration of Translation Medicine paradigm (see Figure 1) in the drug discovery process.

  • Evolution of the physicochemical properties of marketed drugs: can history foretell the future?. Faller B. et al. Drug Discov Today. (5375)
    Involved Partner: NOVARTIS
  • KEYWORDS: ADME – DATABASES – MOLECULAR DESCRIPTORS

    Discussion of the evolution of those required physicochemical properties (high passive permeability, moderate plasma protein binding and low efflux in the Caco-2 permeability assay) for marketed drugs criteria, which are currently pronunced to be changed in the ‘coming’ drug property space. A subset of 12,000 bioactive small molecules from the GVK MedChem database with 400,000 molecules extracted from literature have been considered to evaluate their physicochemical properties (see Table 1 for details of the 56 ChemMAP descriptors used) coverage.

  • Prediction model of potential hepatocarcinogenicity of rat hepatocarcinogens using a large-scale toxicogenomics database. Uehara T. et al. Toxicol Appl Pharmacol. (5373)
  • KEYWORDS: ASSAY DATA – CARCINOGENICITY – HEPATOTOXICITY – TOXICOGENOMICS

    The set of 150 compounds gathered in the database TG-GATEs (Genomics-Assisted Toxicity Evaluation System) have been considered to develop a robust gene-based (Abcb1a/1b, Cd276, Ica1, Tmem184c, Acot9, Tes and Cdh13) model to predict genotoxic or non-genotoxic potential of hepatocarcinogens based on data of 28 day study in rats (see Table 1 for chemicals names, and Support Material Table 1 for detailed experimental conditions).

  • Crystal structure of the β2 adrenergic receptor–Gs protein complex. Rasmussen SG. et al. Nature. (5372)
  • KEYWORDS: PHARMACOLOGY

    Article to report the structural featurs of the β2AR–Gs complex (PDBe:3sn6, ligand data is at CHEMBL:1615159) which was crystallized from β2AR and Gs protein expressed in insect cells.

  • Predicting Drug-induced Hepatotoxicity Using QSAR and Toxicogenomics Approaches. Low Y. et al. Chem Res Toxicol. (5371)
  • KEYWORDS: ASSAY DATA – DILI – HEPATOTOXICITY – MOLECULAR DESCRIPTORS – QSAR MODELING – TOXICOGENOMICS

    Based on data from a 28-day study on rats for a set of 127 drugs (chemical features plus toxicogenomic data; see support material for chemical names, CAS number, dosage, administration route and vehicle, histopathology , serum chemistry (ALT, AST, ALP, TBIL, DBIL and GGT enzymes), list of predictive gene biomarkers and the involving pathways, and list of compounds paired by chemical and transcriptional similarities information), the resulting model generated evidences the enrichment of using different descriptors and toxicogenomics profiles to tackle drug-induced hepatotoxicity prediction.

  • Structural Alert/Reactive Metabolite Concept as Applied in Medicinal Chemistry to Mitigate the Risk of Idiosyncratic Drug Toxicity: A Perspective Based on the Critical Examination of Trends in the Top 200 Drugs Marketed in the United States. Stepan AF. et al. Chem Res Toxicol. (5370)
    Involved Partner: PFIZER
  • KEYWORDS: ASSAY DATA – DRUG SAFETY – METABOLISM – STRUCTURE-BASED PREDICTION

    A comprehensive overview to discuss different analysis (physicochemical trends, structural alerts and reactive metabolite formation) to strategically mitigate the risk of idiosyncratic drug toxicity based on a set of the top 200 drugs marketed in the United States (see tables 1-4 where information on drugs indication, acid/basic/neutral feature, logP, TPSA, reason for withdrawal, daly dose, alerst and positive reactive metabolite formation data are provided).

  • Lead detoxification activities and ADMET hepatotoxicities of a class of novel 5-(1-carbonyl-L-amino-acid)-2,2-dimethyl-[1,3]dithiolane-4-carboxylic acids. Xu Y. et al. Bioorg Med Chem Lett. (5368)
  • KEYWORDS: ADME – ASSAY DATA – HEPATOTOXICITY

    This article presents ADME data regarding a novel series of 12 compounds that thanks to their transmembrane ability are effective in lowering the organs’ lead level of the mice (femurs, brain and kidney) and reports in silico prediction of their non-hepatotoxin character (see Table 1 for values of lead in organs of the treated mice, tables 3 and 4 for permeability coefficients and ADME hepatotoxicity probability, respectively; and support material for experimental details and physical chemical constants).

  • The intra- and inter-laboratory reproducibility and predictivity of the KeratinoSens assay to predict skin sensitizers in vitro: results of a ring-study in five laboratories. Andreas N. et al. Toxicol In Vitro. (5364)
  • KEYWORDS: ASSAY DATA – SKIN SENSITIZATION

    A successful initiative to evaluate intralaboratory reproducibility of skin sensitization prediction is presented and shows the easy transferability of data between 5 different labs and the agreement in prediction for a set of 28 compounds tested (see tables 1 and 2, and support material that provide their CAS number, commercial source and batch, and sensitization potential determined by the Local Lymph Node Assay).

  • Comparative Analysis of Predictive Models for Nongenotoxic Hepatocarcinogenicity Using Both Toxicogenomics and Quantitative Structure–Activity Relationships. Liu Z. et al. Chem Res Toxicol. (5361)
  • KEYWORDS: CARCINOGENESIS – HEPATOTOXICITY – QSAR MODELING – TOXICOGENOMICS

    Data of 5288 microarrays associated with animals treated with 344 chemicals from Gene Expression Omnibus database has been considered to compare results of predictivity taking into account toxicogenomics data (see information in Table 2) or applying QSAR methodology (see Table 1 for chemical names, CAS numbers, dosage and classification).

  • Metabolism and Biomarkers of Heterocyclic Aromatic Amines in Molecular Epidemiology Studies: Lessons Learned from Aromatic Amines. Turesky RJ. et al. Chem Res Toxicol. (5360)
  • KEYWORDS: BIOMARKERS – CYP450 – ENZYMES – METABOLISM – RISK ASSESSMENT

    A comprehensive review on Aromatic Amines metabolism and biomarkers related information currently available in the field of drug safety. Examples are presented to cover different biotransformation reactions that these structures related with carcinogens classes might experience during their metabolism process (CYP450 and Phase II enzymes activity). Discussion on long-term biomarkers is also reported to stress the critical need to establish a list of them to be included in epidemiology studies for an early risk assessment.

  • FDA-approved drug labeling for the study of drug-induced liver injury. Chen M. et al. Drug Discov Today. (5347)
  • KEYWORDS: DILI – HEPATOTOXICITY – REGULATORY GUIDELINES – TEXT MINING

    Based on the FDA benchmark dataset of 287 drugs (see Support Material for drugs related information) which represents a diverse collection of therapeutic categories and a fully covered dosage range (see Figure 3), the authors propose a systematic classification scheme applying FDA-approved drug labeling to assess the Drug-induced liver injury.

  • Elucidation of Common Pharmacophores from Analysis of Targeted Metabolites Transported by the Multispecific Drug Transporter—Organic Anion Transporter1 (Oat1). Kouznetsova VL. et al. Bioorg Med Chem. (5338)
  • KEYWORDS: DATABASES – METABOLISM – PHARMACOLOGY – STRUCTURED- BASED PREDICTION – TRANSPORTERS

    Four pharmacophore hypotheses were designed based on compounds annotated to the multiple drug transporter-Organic anion transporter (Oat1) in the Open NCI database, and their analysis highlighted the common functional groups features of this set: a common COO- group-a doble oxygen (O2) center- or a COOH group-an anionic group, both with acceptor character; and also at least 2 hydrophobic centers, a H-bond donor and a H-bond acceptor. Screening of a commercial database provided by CCG (653,214 compounds) resulted in 12,713 hits.

June

May

  • Comparison of a genomic and a multiplex cell imaging approach for the detection of phospholipidosis. Tilmant K. et al. Toxicol In Vitro. (5337)
    Involved Partner: UCB
  • KEYWORDS: ASSAY DATA – PHOSPHOLIPIDOSIS – RISK ASSESSMENT

    Based on genomic and cell imagining approaches application, authors review (see Table 1) a set of 26 known compounds regarding phospholipidosis activity (15 inducers and 7 non-inducers) and additionaly, a set of 11 UCB propietary compounds (see Table 2 for data obtained from in vivo biological findings in a 4-day exploratory rat toxicity study). Comparison of sensitivity and specificity between both approaches evidences that the fluorescent method seems to be the most appropiate method to predict this toxicity event.

  • Qualitative pharmacology in a quantitative world: diminishing value in the drug discovery process. Williams M. Curr Opinion in Pharmacol. (5334)
  • KEYWORDS: DRUG DISCOVERY

    An brief overview of the general features of new chemical entities (NCE) (see Table 1 for a general outline).

  • Definition of Drug-Likeness for Compound Affinity. Fukunishi Y. et al. J Chem Inf Model. (5333)
  • KEYWORDS: DRUG DISCOVERY – PHARMACOLOGY

    The evaluation of a set of compounds against two sets of proteins (A=160 and B=53, see appendixes A an B) that not include target proteins of the used active compounds shows how the deviation of docking scores within a set of proteins could be a way to measure the drug-likeness for compounds affinity.

April

  • Challenges for the prediction of macromolecular interactions. Wass MN. et al. Curr Opin Struct Biol. (5331)
  • KEYWORDS: DRUG DISCOVERY – PHARMACOLOGY

    This article faces the conservation of interfaces and ligand binding sites trying to answer questions like i) Given a protein can we predict its ligand binding site?, ii) For a proteome can we predict its interactome? and iii) For interacting proteins can we predict their interfaces and the structure of the complex they form?; with a deep discussion around prediction of protein-protein interactions, prediction of protein-protein interaction sites and prediction of ligand binding sites.

  • Classification of Cytochrome P450 Inhibitors and Noninhibitors Using Combined Classifiers. Cheng F. et al. J Chem Inf Model. (5330)
  • KEYWORDS: ASSAY DATA – CYP450 – DATABASES – PHARMACOLOGY

    Based on the largest data set of over 24700 unique compounds from PubChem, this article reports the combined classifiers models to predict inhibition of 1A2, 2C9, 2C19, 2D6 and 3A4 cytochromes (see Figure 2 for the whole work-flow for combined classifiers models building and validation and find data of the data set in their own website, Lmmd).

  • Novel Method for Pharmacophore Analysis by Examining the Joint Pharmacophore Space. Ranu S. et al. J Chem Inf Model. (5327)
  • KEYWORDS: DATABASES – DATA MINING – MOLECULAR DESCRIPTORS – PHARMACOLOGY – STRUCTURE-BASED

    From a new geometric perspective, this article provides a unique definition of a Joint Pharmacophore space (JPS) based on use of 3D geometry of pharmacophoric features for a set of actives against multiple targets; and demonstrates analyzing the problem of pharmacophore modeling on 32 datasets obtained from different sources (11 anticancer screen data sets (breast, 2 leukemia, non-small cell lung, ovarian, prostate, central nervous system, renal, colon, melanoma and yest anticancer) from PubChem, 10000 molecules assayed against cyclin-dependent kinase 5 (CDK-5) and 20 data sets from DUD) how the identification of subspaces has novel data mining applications (see figures 4 and 6 for an outline of the proposed approach).

  • Assessment of Methods To Define the Applicability Domain of Structural Alert Models. Ellison CM. et al. J Chem Inf Model. (5326)
    Involved Partners: LJMU, LL
  • KEYWORDS: MOLECULAR DESCRIPTORS – MUTAGENICITY – QSAR MODELLING – RISK ASSESSMENT – SOFTWARE – STRUCTURE-BASED

    Based on publicly available Ames data of 4337 compounds (2401mutagens and 1936 nonmutagens), this article presents the examination of possible methods for the applicability domain definition of structural alerts models (fragment-based approach, chemical descriptor ranges, structural similarity and specific applicability domain definition (AMBIT discovery program)). Results evidence that the use of a combination of different techniques allows different factors to have a role in the definition of the limits of the domain and for those compounds correctly predicted that are found within the applicability domain are presumably to be correctly predicted than those found outside of a certain domain. This study was performed with the application of the knowledge-based predictive toxicology program, Derek for Windows.

  • The Molecular Informatics journal presents an special issue with articles from the ‘18th European Symposium on Quantitative Structure-Activity Relationships’ (EuroQSAR 2010), celebrated in Rhodes, Greece, 19-24 September 2010.
  • Managing the challenge of chemically reactive metabolites in drug development. Park BK. et al. Nat Rev Drug Discov. (5323)
    Involved Partners: GSK, AZ, PFIZER, BI, VUA
  • KEYWORDS: DRUG SAFETY – METABOLISM – PHARMACOLOGY

    Medicinal chemists, toxicologists and clinicians need drug metabolism information from the chemical reactive metabolites (CRM) to assess drug safety along the drug development process. This review refers first to existing experimental detection methods (measurement of a drug that becomes irreversibly bound to a protein, in either in vitro or animal studies, and mass spectrometric detection of thioether adducts and/or conjugates, see Box 1); presents the lessons taken from structural alerts highly frequently associated with severe toxicities (anilines and anilides; arylacetic and arylpropionic acids; hydrazines and hydrazides; thiophenes; nitroaromatics; and structures that either contain or form α,β-unsaturated enal and/or enone-like structures, including quinones and quinone methides, see Box 2); discusses about physiological response to drug bioactivation (see Box 3) and drug metabolism and hypersensitivity (see Box 4) based on different pharma companies case studies and lists identified unsolved problems and paradoxes. Attention to figures 1 and 2 is recommended for those who afford the relationship between chemistry and drug metabolism and look for examples of possible decision scheme structures to handle bioactivation data along the drug research, respectively.

  • The resurgence of covalent drugs. Singh J. et al. Nat Rev Drug Discov. (5322)
  • KEYWORDS: ADME – DRUG DISCOVERY – METABOLISM – PHARMACOLOGY

    A review focused in the limitations and advantages of covalent drugs in terms of related safety, their mechanistic and pharmacological features (potency, selectivity and pharmacodynamics, see Figure 3 for a list of special considerations in the discovery and development of targeted covalent inhibitors) and their resistance. Examples of covalent drugs in current clinical development (see Table 1) are provided to evidence the recent emerging attention to this therapeutic class of drugs thanks basically the coupling between structural bioinformatics approaches and structure-based drug design.

  • The value of data. Mons B. et al. Nat Genet. (5321)
  • KEYWORDS: DATABASES – DATA MINING – NLP – TEXT MINING

    This commentary article points out the current needs of innovative ways of data sharing and review some initiatives like Open PHACTS project that aims to create an Open Pharmacological Space. Authors discuss about the state of scholarly communication and remark that more than 20 millions articles exist in biomedicine alone. Strategies to link and aggregate them should be addressed if data contents is presented as machine-readable and stored in an unambiguous format.

March

  • Toward a new age of cellular pharmacokinetics in drug discovery. Zhou F. et al. Drug Metab Rev. (5315)
  • KEYWORDS: ADME – CYP450 – ENZYMES – METABOLISM – PHARMACOKINETICS – PHARMACOLOGY – TRANSPORTERS

    A discussion of several factors (transport, disposition, metabolism, pH partitioning or electrochemical gradient) that can have influence in the cellular (hepatocytes, tumor cells or enterocytes) pharmacokinetics of drugs (see Table 1 for a representative list of drugs that their cellular pharmacokinetics are determinant ot their pharmacological responses).

  • Capturing Structure−Activity Relationships from Chemogenomic Spaces. Wendt B. et al. J Chem Inf Model. (5311)
    Involved Partner: EMBL/EBI
  • KEYWORDS: DATABASES – PHARMACOLOGY – QSAR MODELLING

    Based on 250 market drugs (smiles in table S1 of support material) active against one of 22 different therapeutic targets, seven public databases (see Table 1) were consulted to construct SAR tables by assembling similar structures around each query structure that had annotation a certain target and consequent application of Quantitative series enrichment analysis (QSEA) provides a easier identification of selectivity trends.

  • In silico repositioning of approved drugs for rare and neglected diseases. Ekins S. et al. Drug Discov Today. (5309)
  • KEYWORDS: DATABASES – DRUG DISCOVERY

    This review claims to rectify the not widely use for repositioning the FDA-approval drugs presenting discussion around examples of approved drug molecules identified using low-throughput (see Table 1), using HTS or in silico (see Table 2) screening methods as having effects against diseases other than original targets, or using an in silico-in vitro approach to inhibit transporters (see Table 3). Evidences mostly found by machine learning literature analysis and integration of data from databases and in vitro screening. Authors report a list of data sets of antimalarial and tubercolosis databases publicly available from the Collaborative Drug Discovery database (CDD, see Table 4).

  • The role of translational bioinformatics in drug discovery. Buchan NS. et al.Drug Discov Today. (5308)
  • KEYWORDS:BIOMARKERS – DATABASES – DRUG DISCOVERY

    A review about the current impact of translational bioinformatics approaches on preclinical and clinical research, specially in fields like drug repurposing or drug safety, is presented together with discussion of several examples of data sources where translational bioinformatics analyses are applied (see Table 1 for a list of molecular, health record, clinical trial, safety data, drug and drug target and claim data resources).

  • QSAR-based permeability model for drug-like compounds. Gozalbes R. et al. Bioorg Med Chem. (5307)
  • KEYWORDS: ADME – ASSAY DATA – MOLECULAR DESCRIPTORS – STRUCTURE-BASED PREDICTION – QSAR MODELLING

    Based on publicaly available Caco-2 permeability data, a new QSAR permeability model was generated trying to overcome the limitations of previous models of human passive intestinal absorption for potential drugs. This article provides experimental and predicted permeability values for the list of training set compounds (see Table 1), the molecules extracted from the FDA Biopharmaceutics classification system (BCS) (see Table 3) and the 21 drug-like compounds used as external set.

  • Clinical importance of OATP1B1 and OATP1B3 in drug-drug interactions. Shitara Y. Drug Metab Pharmacokinet. (5302)
  • KEYWORDS: ADME – ASSAY DATA – CYP450 – DRUG-DRUG INTERACTIONS – PHARMACOKINETICS – TRANSPORTERS

    Analysis of different drugs coadministration adverse effects (cycloporin A or gemfibrozil together with statins, see tables 3 and 4) is presented by evaluation of pharmacokinetic data (AUC, Cmax, t1/2) and reasons are found to have relation with the inhibition or induction of transporters like OATP1B1 and OATP1B3 expressed in the liver. Additionally, the effects of genetic polymorphisms in pharmacokinetic parameters alteration is reported for different substrates of OATP1B1(atorvastatin, pravastatin, rosuvastatin, nateglinide and repaglinide, see tables 1 and 2).

  • Databases in the area of pharmacogenetics. Sim SC. et al. Hum Mutat. (5301)
  • KEYWORDS: CYP450 – DATABASES – ENZYMES – PHARMACOLOGY – TRANSPORTERS

    Review of different databases in the pharmacogenomics field: the Pharmacogenomics Knowledge Base (PharmGKB), the Human Cytochrome P450 (CYP) Allele Nomenclature website, the Human Arylamine N-Acetyltransferase (NAT) Gene Nomenclature, the Pharmacogenetics of Membrane Transporters (PMT) database, Transporter database (TP-search), UGT Alleles Nomenclature page, and PharmaADME.

  • Probing small-molecule binding to cytochrome P450 2D6 and 2C9: An in silico protocol for generating toxicity alerts. Rossato G. et al. ChemMedChem. (5300)
  • KEYWORDS: CYP450 – METABOLISM – PHARMACOLOGY – QSAR MODELING – STRUCTURE-BASED PREDICTION

    This article presents how a prealignment of a 4D ensemble of 56 compounds for CYP2D6 and 85 for CYP2C9 (see structures in Support Material section) improves sensitivity in the generation of multidimensional QSAR models (software Quasar applied, see table 2 and 3 where information of external set used for validation is provided). The results of this study underline the fact that for the correct modeling of drug-like molecules binding to these CYP2D6 and CYP2C9, one must consider ligand–pharmacophore features, state of oxidation of the heme, and active-site motion. Note: both models are available in the VirtualToxLab.

  • Impact of high-throughput screening in biomedical research. Macarron R. et al. Nat Rev Drug Discov. (5296)
    Involved Partners: SAD, GSK, PFIZER, NOVARTIS
  • KEYWORDS: DRUG DISCOVERY – PHARMACOLOGY

    This opinion article discusses the integration of high-throughput screening as part of the drug discovery process in both pharmaceutical and biomedical research. Pros and cons of this methodology are reported regarding quality of data generated, the costs and time consuming, its lack of anti-intellectual and irrational application, the optimal composition of a screening collection and also their role in the evolution of compounds libraries. Examples of recently approved drugs with origins in HTS hits (see Table 2) and recent initiatives of academic institutions establishing HTS facilities seem to enhance its impact in drug discovery.

  • Probing the links between in vitro potency, ADMET and physicochemical parameters. Gleeson MP. et al. Nat Rev Drug Discov. (5295)
    Involved Partners: EMBL/EBI, GSK
  • KEYWORDS: ADME – DATABASE – DRUG DISCOVERY – MOLECULAR DESCRIPTORS – PHARMACOLOGY

    This review provides a range of analysis (see support material) based on data from ChEMBL database (see Table 1 for a distribution of physicochemical properties comparison between ChEMBL and a GSK oral drugs set) to highlight probable relationship between ADMET parameters (solubility, permeability, bioavailability and cytochrome P450 3A4 inhibition) or in vitro potency respect to physicochemical properties (molecular mass, logP), and also between in vitro potency and therapeutic dose. The results suggest that traditional attempts to filter candidates firstly considering their affinity for a range of screened targets should evolve to firstly evaluation and understanding of their ADMET properties.

  • Genotoxicity and carcinogenicity studies of antihistamines. Brambilla G. et al. Arch Toxicol. (5294)
  • KEYWORDS: ASSAY DATA – CARCINOGENICITY – GENOTOXICITY – REGULATORY GUIDELINES – RISK ASSESSMENT

    An exhaustive search of genotoxicity and carcinogenicity published data shows how data available from 29 of a list of 70 anthistamine marketed drugs (see Table 1 for data related) evidences current needs of adequate evaluation to assess possible genotoxic and carcinogenic activity risk because most of the current accepted drugs have not all data required by present guidelines.

  • Chemogenomic Analysis of G-Protein Coupled Receptors and Their Ligands Deciphers Locks and Keys Governing Diverse Aspects of Signalling. Wichard JD. et al. PLoS One. (5292)
    Involved Partner: BSP
  • KEYWORDS: DATABASES – DRUG DISCOVERY – MOLECULAR DESCRIPTORS – PHARMACOLOGY

    Based on a collection of 1644 receptor-ligand pairs information covering 100 family A GPCRs gathered from IUPHAR database and extraction of their 2D chemical structures from the PubChem Structure Database, an analysis of antagonist and agonist pharmacological patterns identifies hot-spots of correlated GPCR residue positions and specific ligand properties by separating antagonistic and agonistic effects on signaling (see Figure 2, and corresponding mapping to the rhodopsin crystal structure (see Figure 3)). Additionally, evaluation of more than 30 ligand descriptors like number of H-donors, H attached to heteroatom or number of ring bonds highlights the most significant ligand descriptors (see Figure 4 for a list of descriptors sorted according to their frequency of occurrence); where most of them are found for both types of ligands, agonistic and antagonistic, and few shared descriptors are found in high or low amount for agonists or antagonists, respectively.

February

  • Formation of mechanistic categories and local models to facilitate the prediction of toxicity. Cronin MT. et al. ALTEX. (5288)
    Involved Partner: LJMU
  • KEYWORDS: DRUG SAFETY – QSAR MODELING – RISK ASSESSMENT – SKIN SENSITIZATION – TERATOGENICITY

    A brief overview of the state of the art of methods to form local QSAR models through the need to develop chemical categories. Discussion about the strengths and limitations of the chemical categories lists advantages and disadvantages of this methodology. Two different toxicity events are presented, skin sensitization and teratogenicity, as case studies of two different criteria of chemical categories formation, basis of mechanism of action and basis of chemical similarity, respectively.

  • From alternative methods to a new toxicology. Hartung T. Eur J Pharm Biopharm. (5287)
  • KEYWORDS: DRUG DISCOVERY – REGULATORY GUIDLEINES

    This review stresses on a change for a paradigm shift in the toxicology study field. From legislation issues (animal welfare legislation, EU cosmetic legislation and REACH and TSCA reauthorization) to limitations of current tools for prediction are discussed to remark needs of new tests (in vivo, in vitro or in silico) and to face the emerging solutions like integrated testing strategies, application of new technologies, or probabilistic assessment.

  • How can we improve our understanding of cardiovascular safety liabilities to develop safer medicines?. Laverty HG. et al. Br J Pharmacol. (5286)
    Involved Partners: AZ, GSK, SAD, NOVARTIS
  • KEYWORDS: CARDIOTOXICITY – DRUG SAFETY

    Report on the Medical Research Council Centre for Drug Safety Science workshop on ‘Cardiovascular Toxicity Medicines’. This article summarises the key discussions relating to cardiovascular safety liabilities to understand cause of drug attrition during pre-clinical and clinical development, adverse drug reactions and post-approval withdrawal of medicines (see tables 1, 2 and 3 to have disclosed information depending the tissue effected (heart, vessel and blood)). Authors provide a compilation of current key initiatives focused on cardiovascular safety liabilities (see Table 4).

  • Fragment screening to predict druggability (ligandability) and lead discovery success. Edfeldt FNB. et al. Drug Discov Today. (5285)
    Involved Partner: AZ
  • KEYWORDS: DRUG DISCOVERY – PHARMACOLOGY

    This article remarks ligandability definition versus druggability concept as new perspective to have into account in fragment screening methodologies. Authors quantify the power of fragment screening by classification of data from 36 discovery projects from period 2001-2008 about 36 targets (see Table 1) in three categories high/medium/low ligandability based on affinities values < 0.1 mM/0.1-1.0 M/>1 M and larger/some/low diversity, respectively. Application of such methodology could provide an early ‘fingerprint’ of the physicochemical property space and advance lead-like compounds previously to the hit-to-lead stage.

  • Volume to dissolve applied dose (VDAD) and apparent dissolution rate (ADR) – tools to predict in vivo bioavailability from orally applied drug suspensions. Muenster U. et al. Eur J Pharm Biopharm. (5284)
  • KEYWORDS: ADME – ASSAY DATA – PHARMACOKINETICS

    Analysis of physicochemical and pharmacokinetic properties data of 37 compounds (see Table 1, without disclosing their structures due to intellectual property reasons) reveals that the apparent dissolution rate (ADR) and the volume to dissolve applied dose (VDAD) need to achieve Frel=50% in rat to assure a sufficient in vivo dissolution in human after oral application of solid formulations, and drugs should exhibit a VDAD of aprox 100 (pH 4.5)-500 (pH 7) ml/kg or less in aqueous media to avoid insufficient or varying drug absorption due to dissolution limitation.

  • Nuclear Receptor PXR, transcriptional circuits and metabolic relevance. Ihunnah CA. et al. Biochim Biophys Acta. (5283)
  • KEYWORDS: CYP450 – DRUG-DRUG INTERACTION – ENZYMES – METABOLISM – NUCLEAR RECEPTORS – PHARMACOLOGY – TRANSPORTERS

    A brief review on PXR nuclear receptor action in xenobiotic metabolism (regulation of Phase I and Phase II enzymes, transporters and implication in drug-drug interactions) and in endobiotic metabolism (role in glucose metabolism, lipid metabolism, glucocorticoid and mineralocorticoid homeostasis, androgen metabolism, bile acid and bilirubin detoxification, vitamin metabolism and bone mineral homeostasis, retinoic acid metabolism and inflammation).

  • A core in vitro genotoxicity battery comprising the Ames test plus the in vitro micronucleus test is sufficient to detect rodent carcinogens and in vivo genotoxins. Kirkland D. et al. Mutat Res. (5282)
  • KEYWORDS: ASSAY DATA – GENOTOXICITY

    A comprehensive set of 962 rodent carcinogens with in vitro data (micronucleus test (MNvit) and chromosomal aberration test (CAvit)) and in vivo data (mouse-lymphoma assay (MLA)) is evaluated and its analysis suggests that no relevant rodent carcinogens or in vivo genotoxins would remain undetected in an in vitro baterry combining both Ames and MNvit tests information (see Appendix A for all experimental data known for these compounds).

  • Drug discovery: A question of library design. Hajduk PJ. et al. Nature. (5279)
  • KEYWORDS: DATABASES – DRUG DESIGN

    Briefly, this forum of drug discovery discussions advocates to two approaches for creating libraries of compounds (fragment-based ligand design and diversity-oriented synthesis) for use in biological screening assays for drug discovery and manifest pros and cons to find better leads for medicinal chemistry programmes.

  • Computational Polypharmacology with Text Mining and Ontologies. Plake C. et al. Curr Pharm Biotechnol. (5278)
  • KEYWORDS: DATABASES – PHARMACOLOGY – TEXT MINING

    A simple analysis to highligth the potential of applying text-mining approaches (see Table 3 for a list of studies published) and ontologies (see Table 1 for a list of databases consulted to have concepts to describe drugs and targets) to identify target and drug-target interaction from selected web resources (see Table 2).

  • Activity profiles of 309 ToxCast™ chemicals evaluated across 292 biochemical targets. Knudsen TB. et al. Toxicology. (5277)
  • Recent example of cross-pharmacology analysis to understand the susceptible role of different targets that may lead to key events in toxicology. The 309 compounds currently gathered in the phase-I of ToxCast chemical library were evaluated across 292 biochemical: GPCRs, CYP450, kinases, phosphatases, proteases, HDACs, nuclear receptors, ion channels, and transporters (see Figure 1 and support material for compound-target relationship). The results confirm known targets and alert about novel targets related to each compound case.

  • New uses for old drugs: pharmacophore-based screening for the discovery of P-glycoprotein inhibitors. Palmeira A. et al.Chem Bio Drug Des. (5276)
  • KEYWORDS: ASSAY DATA – P-gp – PHARMACOLOGY – TRANSPORTERS

    In this article, a selection of 26 known P-gp inhibitors are considered to build a pharmacophore model (using PharmaGist, Corina, and VlifeMDS) to screen the Drug Bank database. The 21 hits are evaluated experimentaly by a rhodamine-123 accumulation assay using K562Dox cell line to discriminate between activatiors (9) and inhibitors (12) and this latter group between non-competitive (6) and competitive (6) by a P-gp ATPase activity assay. Additionally,a PCA analysis evidences a linear combination of descriptors and the hydrophobicity and the number of hydrogen bond donor and acceptor for those molecules that interact with P-gp.

January

  • Are circulating metabolites important in drug-drug interactions?: Quantitative analysis of risk prediction and inhibitory potency. Yeung CK. et al. Clin Pharmacol Ther. (5275)
  • KEYWORDS: ASSAY DATA – CYP450 – METABOLISM – PHARMACOLOGY

    Based on 24 drugs and their corresponding metabolites data (see Table 1), this article discusses about the relative importance of those circulating metabolites that may contribute to specific in vivo drug-drug interactions to better understand some drug actions and toxicity occurrence.

  • Automated Information Extraction and Structure−Activity Relationship Analysis of Cytochrome P450 Substrate. Yamashita F. et al. J Chem Inf Model. (5271)
  • KEYWORDS: CYP450 – NLP – PHARMACOLOGY – TEXT MINING

    A comprehensive set of article references regarding CYP1A2, CYP2C9, CYP2C19, CYP2D6, CYP2E1 and CYP3A4 substrates, inhibitors and inducers (see Support information) is presented as result of text-mining technique application using natural language processing. Authors provide an example of the interaction extraction process used (see Table 1), and the list of verbs and verbal nouns used for classification of different chemicals identified from literature into substrates, inhibitors and inducers (see Table 2).

  • Bioactivation of Drugs: Risk and Drug Design. Walsh JS. et al. Annu Rev Pharmacol Toxicol. (5270)
  • KEYWORDS: CYP450 – METABOLISM – PHARMACOLOGY

    An overview of the detected key issues for bioactivation through drug metabolism with special focus on structural alerts (see tables 1 and 2) and different strategies commonly used to minimize bioactivation risk (blocking bioactivation of activated aromatic systems and five-membered heterocycles; metabolic switching, reduction of oxidation potential to make metabolism less favorable, or the case of fluorination technique to block metabolism).

  • Combined Receptor and Ligand-Based Approach to the Universal Pharmacophore Model Development for Studies of Drug Blockade to the hERG1 Pore Domain. Durdagi S. et al. J Chem Inf Model. (5268)
  • KEYWORDS: ASSAY DATA – hERG – PHARMACOLOGY – STRUCTURE-BASED PREDICTION

    A pharmacophore model is presented based on a set of 31 drugs, validated with a set of 9 compounds and applied to predict hERG activity of a series of 14 compounds as additional external set (see tables 1 and 2, and also TS2 in the Support information document for names, structure picture and activity values, respectively).

  • Phthalates: Toxicogenomics and inferred human diseases. Singh S. et al. Genomics. (5263)
  • KEYWORDS: ASSAY DATA – CARDIOTOXICITY – CYP450 – DATABASES – HEPATOTOXICITY – NEPHROTOXICITY – NUCLEAR RECEPTORS – PHARMACOLOGY – PATHWAYS – SYSTEMS BIOLOGY – TOXICOGENOMICS

    Focused on a group of 16 phthalates, the Comparative Toxicogenomics Database (CTD) is consulted to identify interactions between genes/proteins and these compounds (see Table 1 for a list of compounds evaluated, Table 2 for the list of the 34 most frequent genes/proteins connected to them (basically Nuclear Receptors, CYP450s, GPCRs and Ion channels). Analysis of these results suggests that Cardiotoxicity, Hepatotoxicity and Neprhotoxicity are expected to be the top toxicity categories for this series of compounds and points out the need to check this list of proteins as a potential biomarkers of phthalate toxicity.

  • Predicting hERG activities of compounds from their 3D structures: Development and evaluation of a global descriptors based QSAR model. Sinha N. et al. Eur J Med Chem. (5260)
  • KEYWORDS: ASSAY DATA – hERG – MOLECULAR DESCRIPTORS – QSAR MODELING

    Based on a training set of 77 compounds covering a wide range of activities (see Table 1, smiles and experimental pIC50 values), a external test set of 80 compounds (see Table 2) and an additional set of 32 compounds (see Table 4) are evaluated and the results evidence that different types of binding modes may explain different ways to block hERG and then a single QSAR cannot predict activities for this wide range of structural diversity. See Support information for compound structure pictures and data.

  • Evaluation of the OECD QSAR Application Toolbox and Toxtree for estimating the mutagenicity of chemicals. Part 1. Aromatic amines. Devillers J. et al. SAR QSAR Environ Res. (5259)
  • KEYWORDS: ASSAY DATA – MUTAGENICITY – QSAR MODELING – REGULATORY GUIDELINES

    An attempt to evaluate both OECD QSAR application Toolbox and Toxtree models for mutagenicity prediction questions, in the case of a set of 152 aromatic amines used as external test set, the prediction potential of such referential models. Authors provide data (name, CAS number and Observed and Calculated mutagenic behaviour) for the 18 chemicals of the training set (see Table 1) and for the 152 chemicals (see Table 2, also Toolbox and Toxtree behaviour and structural alerts predictions are included).