Mol2vec: Unsupervised Machine Learning Approach with Chemical Intuition.

Inspired by natural language processing techniques, we here introduce Mol2vec, which is an unsupervised machine learning approach to learn vector representations of molecular substructures. Like the Word2vec models, where vectors of closely related words are in close proximity in the vector space, Mol2vec learns vector representations of molecular substructures that point in similar directions for chemically related substructures. Compounds can finally be encoded as vectors by summing the vectors of the individual substructures and, for instance, be fed into supervised machine learning approaches to predict compound properties. The underlying substructure vector embeddings are obtained by training an unsupervised machine learning approach on a so-called corpus of compounds that consists of all available chemical matter. The resulting Mol2vec model is pretrained once, yields dense vector representations, and overcomes drawbacks of common compound feature representations such as sparseness and bit collisions. The prediction capabilities are demonstrated on several compound property and bioactivity data sets and compared with results obtained for Morgan fingerprints as a reference compound representation. Mol2vec can be easily combined with ProtVec, which employs the same Word2vec concept on protein sequences, resulting in a proteochemometric approach that is alignment-independent and thus can also be easily used for proteins with low sequence similarities.

KinMap: a web-based tool for interactive navigation through human kinome data.

BACKGROUND: Annotations of the phylogenetic tree of the human kinome is an intuitive way to visualize compound profiling data, structural features of kinases or functional relationships within this important class of proteins. The increasing volume and complexity of kinase-related data underlines the need for a tool that enables complex queries pertaining to kinase disease involvement and potential therapeutic uses of kinase inhibitors. RESULTS: Here, we present KinMap, a user-friendly online tool that facilitates the interactive navigation through kinase knowledge by linking biochemical, structural, and disease association data to the human kinome tree. To this end, preprocessed data from freely-available sources, such as ChEMBL, the Protein Data Bank, and the Center for Therapeutic Target Validation platform are integrated into KinMap and can easily be complemented by proprietary data. The value of KinMap will be exemplarily demonstrated for uncovering new therapeutic indications of known kinase inhibitors and for prioritizing kinases for drug development efforts. CONCLUSION: KinMap represents a new generation of kinome tree viewers which facilitates interactive exploration of the human kinome. KinMap enables generation of high-quality annotated images of the human kinome tree as well as exchange of kinome-related data in scientific communications. Furthermore, KinMap supports multiple input and output formats and recognizes alternative kinase names and links them to a unified naming scheme, which makes it a useful tool across different disciplines and applications. A web-service of KinMap is freely available at http://www.kinhub.org/kinmap/ .

Coupling Matched Molecular Pairs with Machine Learning for Virtual Compound Optimization.

Matched molecular pair (MMP) analyses are widely used in compound optimization projects to gain insights into structure-activity relationships (SAR). The analysis is traditionally done via statistical methods but can also be employed together with machine learning (ML) approaches to extrapolate to novel compounds. The here introduced MMP/ML method combines a fragment-based MMP implementation with different machine learning methods to obtain automated SAR decomposition and prediction. To test the prediction capabilities and model transferability, two different compound optimization scenarios were designed: (1) "new fragments" which occurs when exploring new fragments for a defined compound series and (2) "new static core and transformations" which resembles for instance the identification of a new compound series. Very good results were achieved by all employed machine learning methods especially for the new fragments case, but overall deep neural network models performed best, allowing reliable predictions also for the new static core and transformations scenario, where comprehensive SAR knowledge of the compound series is missing. Furthermore, we show that models trained on all available data have a higher generalizability compared to models trained on focused series and can extend beyond chemical space covered in the training data. Thus, coupling MMP with deep neural networks provides a promising approach to make high quality predictions on various data sets and in different compound optimization scenarios.

Identification and Visualization of Kinase-Specific Subpockets.

The identification and design of selective compounds is important for the reduction of unwanted side effects as well as for the development of tool compounds for target validation studies. This is, in particular, true for therapeutically important protein families that possess conserved folds and have numerous members such as kinases. To support the design of selective kinase inhibitors, we developed a novel approach that allows identification of specificity determining subpockets between closely related kinases solely based on their three-dimensional structures. To account for the intrinsic flexibility of the proteins, multiple X-ray structures of the target protein of interest as well as of unwanted off-target(s) are taken into account. The binding pockets of these protein structures are calculated and fused to a combined target and off-target pocket, respectively. Subsequently, shape differences between these two combined pockets are identified via fusion rules. The approach provides a user-friendly visualization of target-specific areas in a binding pocket which should be explored when designing selective compounds. Furthermore, the approach can be easily combined with in silico alanine mutation studies to identify selectivity determining residues. The potential impact of the approach is demonstrated in four retrospective experiments on closely related kinases, i.e., p38α vs Erk2, PAK1 vs PAK4, ITK vs AurA, and BRAF vs VEGFR2. Overall, the presented approach does not require any profiling data for training purposes, provides an intuitive visualization of a large number of protein structures at once, and could also be applied to other target classes.

Pocketome of human kinases: prioritizing the ATP binding sites of (yet) untapped protein kinases for drug discovery.

Protein kinases are involved in a variety of diseases including cancer, inflammation, and autoimmune disorders. Although the development of new kinase inhibitors is a major focus in pharmaceutical research, a large number of kinases remained so far unexplored in drug discovery projects. The selection and assessment of targets is an essential but challenging area. Today, a few thousands of experimentally determined kinase structures are available, covering about half of the human kinome. This large structural source allows guiding the target selection via structure-based druggability prediction approaches such as DoGSiteScorer. Here, a thorough analysis of the ATP pockets of the entire human kinome in the DFG-in state is presented in order to prioritize novel kinase structures for drug discovery projects. For this, all human kinase X-ray structures available in the PDB were collected, and homology models were generated for the missing part of the kinome. DoGSiteScorer was used to calculate geometrical and physicochemical properties of the ATP pockets and to predict the potential of each kinase to be druggable. The results indicate that about 75% of the kinome are in principle druggable. Top ranking structures comprise kinases that are primary targets of known approved drugs but additionally point to so far less explored kinases. The presented analysis provides new insights into the druggability of ATP binding pockets of the entire kinome. We anticipate this comprehensive druggability assessment of protein kinases to be helpful for the community to prioritize so far untapped kinases for drug discovery efforts.

Molecular dynamics to enhance structure-based virtual screening on cathepsin B.

Molecular dynamics (MD) and molecular docking are commonly used to study molecular interactions in drug discovery. Most docking approaches consider proteins as rigid, which can decrease the accuracy of predicted docked poses. Therefore MD simulations can be used prior to docking to add flexibility to proteins. We evaluated the contribution of using MD together with docking in a docking study on human cathepsin B, a well-studied protein involved in numerous pathological processes. Using CHARMM biomolecular simulation program and AutoDock Vina molecular docking program, we found, that short MD simulations significantly improved molecular docking. Our results, expressed with the area under the receiver operating characteristic curves, show an increase in discriminatory power i.e. the ability to discriminate active from inactive compounds of molecular docking, when docking is performed to selected snapshots from MD simulations.

Endocrine disruptome--an open source prediction tool for assessing endocrine disruption potential through nuclear receptor binding.

Predicting the endocrine disruption potential of compounds is a daunting but essential task. Here we report a new tool for this purpose that we have termed Endocrine Disruptome. It is a free and simple-to-use Web service that runs on an open source platform called Docking interface for Target Systems (DoTS). The molecular docking is handled via AutoDock Vina. Compounds are docked to 18 integrated and well-validated crystal structures of 14 different human nuclear receptors: androgen receptor; estrogen receptors α and ß; glucocorticoid receptor; liver X receptors α and ß; mineralocorticoid receptor; peroxisome proliferator activated receptors α, ß/δ, and γ; progesterone receptor; retinoid X receptor α; and thyroid receptors α and ß. Endocrine Disruptome is free of charge and available at http://endocrinedisruptome.ki.si.

Biochemical characterization of MurF from Streptococcus pneumoniae and the identification of a new MurF inhibitor through ligand-based virtual screening.

MurF is an essential bacterial enzyme that is involved in the last intracellular stage of peptidoglycan biosynthesis, and therefore it has the potential to be exploited as a target for the development of new antibacterials. Here, we report on the expression, purification and biochemical characterization of MurF from an important pathogen, Streptococcus pneumoniae. Additionally, ligand-based virtual screening was successfully used and a new hit compound with micromolar inhibitory activities against MurF enzymes from S. pneumoniae and Escherichia coli was identified.

N-Benzoyl anthranilic acid derivatives as selective inhibitors of aldo-keto reductase AKR1C3.

Human aldo-keto reductases AKR1C1-AKR1C3 are involved in the biosynthesis and inactivation of steroid hormones and prostaglandins and thus represent attractive targets for the development of new drugs. We synthesized a series of N-benzoyl anthranilic acid derivatives and tested their inhibitory activity on AKR1C enzymes. Our data show that these derivatives inhibit AKR1C1-AKR1C3 isoforms with low micromolar potency. In addition, five selective inhibitors of AKR1C3 were identified. The most promising inhibitors were compounds 10 and 13, with IC(50) values of 0.31 µM and 0.35 µM for AKR1C3, respectively.

Virtual screening yields inhibitors of novel antifungal drug target, benzoate 4-monooxygenase.

Fungal CYP53 enzymes are highly conserved proteins, involved in phenolic detoxification, and have no homologues in higher eukaryotes, rendering them favorable drug targets. Aiming to discover novel CYP53 inhibitors, we employed two parallel virtual screening protocols and evaluated highest scoring hit compounds by analyzing the spectral binding interactions, by surveying the antifungal activity, and assessing the inhibition of catalytic activity. On the basis of combined results, we selected 3-methyl-4-(1H-pyrrol-1-yl)benzoic acid (compound 2) as the best candidate for hit-to-lead follow-up in the antifungal drug discovery process.

Exploration of the chemical space of novel naphthalene-sulfonamide and anthranilic Acid-based inhibitors of penicillin-binding proteins.

Penicillin-binding proteins are a well established, validated and still a very promising target for the design and development of new antibacterial agents. Based on our previous discovery of several noncovalent small-molecule inhibitor hits for resistant PBPs we decided to additionally explore the chemical space around these compounds. In order to clarify their structure-activity relationships for PBP inhibition two new series of compounds were synthesized, characterized and evaluated biochemically: the derivatives of anthranilic acid and naphthalene-sulfonamide derivatives. The target compounds were tested for their inhibitory activities on three different transpeptidases: PBP2a from methicillin-resistant Staphylococcus aureus (MRSA) strains, PBP5fm from Enterococcus faecium strains, and PBP1b from Streptococcus pneumoniae strains. The most promising results for both of these series of compounds were obtained against the PBP2a enzyme with the IC50 values in the micromolar range. Although these results do not represent a significant breakthrough in the field of noncovalent PBP inhibitors, they do provide useful structure-activity relationship data, and thus a more solid basis for the design of potent and noncovalent inhibitors of resistant PBPs.

Novel inhibitors of trihydroxynaphthalene reductase with antifungal activity identified by ligand-based and structure-based virtual screening.

Curvularia lunata is a dark pigmented fungus that is the causative agent of several diseases in plants and in both immunodeficient and immunocompetent patients. 1,8-Dihydroxynaphthalene-melanin is found in the cell wall of C. lunata and is believed to be the important virulence factor of dematiaceous fungi. Trihydroxynaphthalene reductase is an enzyme of the 1,8-dihydroxynaphthalene-melanin biosynthetic pathway, and it thus represents an emerging target for the development of novel fungicides and antimycotics. In the present study, we describe novel inhibitors of trihydroxynaphthalene reductase from C. lunata. These inhibitors were identified by ligand-based three-dimensional similarity searching and docking to a homology-built model and by subsequent biochemical and antifungal evaluation. Discovery of competitive inhibitors with K(i) values in low micromolar and even nanomolar concentration range proves the aplicability of homology-built model of 3HNR for hit finding by virtual screening methods.

Design and Synthesis of New Peptidomimetics as Potential Inhibitors of MurE.

With the continuing emergence and spread of multidrug-resistant bacteria, there is an urgent need for the development of new antimicrobial agents. One possible source of new antibacterial targets is the biosynthesis of the bacterial cell-wall peptidoglycan. The assembly of the peptide stem is carried out by four essential enzymes, known as the Mur ligases (MurC, D, E and F). We have designed and synthesised a focused library of compounds as potential inhibitors of UDP-N-acetylmuramoyl-L-alanyl-D-glutamate:L-lysine ligase (MurE) from Staphylococcus aureus. This was achieved using two approaches: (i) synthesis of transition-state analogues based on the methyleneamino core; and (ii) synthesis of MurE reaction product analogues. Two methyleneamino-based compounds are identified as initial hits for inhibitors of MurE.

Discovery of new inhibitors of aldo-keto reductase 1C1 by structure-based virtual screening.

Aldo-keto reductase 1C1 is a hydroxysteroid dehydrogenase that inactivates progesterone by converting it to 20alpha-hydroxyprogesterone. It also inactivates 3alpha,5alpha-tetrahydroprogesterone, an allosteric modulator of the gamma-aminobutyric acid receptor that has anaesthetic, analgesic, anxiolytic and anti-convulsant effects. Inhibitors of aldo-keto reductase 1C1 are thus very interesting as potential agents for the treatment of endometrial cancer, premenstrual syndrome, catamenial epilepsy, and depressive disorders, and for the maintenance of pregnancy. We have used the molecular docking program eHiTS for virtual screening of 1990 compounds from the National Cancer Institute "Diversity Set". Fifty compounds with the highest predicted binding energies were then evaluated in vitro. Three structurally diverse hits were obtained that inhibit aldo-keto reductase 1C1 in the low micromolar range of IC(50) values. These hits represent promising starting points for structural optimization in hit-to-lead development.

Design and synthesis of new hydroxyethylamines as inhibitors of D-alanyl-D-lactate ligase (VanA) and D-alanyl-D-alanine ligase (DdlB).

The Van enzymes are ATP-dependant ligases responsible for resistance to vancomycin in Staphylococcus aureus and Enteroccoccus species. The de novo molecular design programme SPROUT was used in conjunction with the X-ray crystal structure of Enterococcus faeciumd-alanyl-d-lactate ligase (VanA) to design new putative inhibitors based on a hydroxyethylamine template. The two best ranked structures were selected and efficient syntheses developed. The inhibitory activities of these molecules were determined on E. faecium VanA, and due to structural similarity and a common reaction mechanism, also on d-Ala-d-Ala ligase (DdlB) from Escherichia coli. The phosphate group attached to the hydroxyl moiety of the hydroxyethylamine isostere within these systems is essential for their inhibitory activity against both VanA and DdlB.

Discovery of new inhibitors of the bacterial peptidoglycan biosynthesis enzymes MurD and MurF by structure-based virtual screening.

The ATP-dependent Mur ligases (MurC, MurD, MurE and MurF) successively add L-Ala, D-Glu, meso-A(2)pm or L-Lys, and D-Ala-D-Ala to the nucleotide precursor UDP-MurNAc, and they represent promising targets for antibacterial drug discovery. We have used the molecular docking programme eHiTS for the virtual screening of 1990 compounds from the National Cancer Institute 'Diversity Set' on MurD and MurF. The 50 top-scoring compounds from screening on each enzyme were selected for experimental biochemical evaluation. Our approach of virtual screening and subsequent in vitro biochemical evaluation of the best ranked compounds has provided four novel MurD inhibitors (best IC(50)=10 microM) and one novel MurF inhibitor (IC(50)=63 microM).

Phosphorylated hydroxyethylamines as novel inhibitors of the bacterial cell wall biosynthesis enzymes MurC to MurF.

Enzymes involved in the biosynthesis of bacterial peptidoglycan represent important targets for development of new antibacterial drugs. Among them, Mur ligases (MurC to MurF) catalyze the formation of the final cytoplasmic precursor UDP-N-acetylmuramyl-pentapeptide from UDP-N-acetylmuramic acid. We present the design, synthesis and biological evaluation of a series of phosphorylated hydroxyethylamines as new type of small-molecule inhibitors of Mur ligases. We show that the phosphate group attached to the hydroxyl moiety of the hydroxyethylamine core is essential for good inhibitory activity. The IC(50) values of these inhibitors were in the micromolar range, which makes them a promising starting point for the development of multiple inhibitors of Mur ligases as potential antibacterial agents. In addition, 1-(4-methoxyphenylsulfonamido)-3-morpholinopropan-2-yl dihydrogen phosphate 7a was discovered as one of the best inhibitors of MurE described so far.

From Cancer to Pain Target by Automated Selectivity Inversion of a Clinical Candidate.

Elimination of inadvertent binding is crucial for inhibitor design targeting conserved protein classes like kinases. Compounds in clinical trials provide a rich source for initiating drug design efforts by exploiting such secondary binding events. Considering both aspects, we shifted the selectivity of tozasertib, originally developed against AurA as cancer target, toward the pain target TrkA. First, selectivity-determining features in binding pockets were identified by fusing interaction grids of several key and off-target conformations. A focused library was subsequently created and prioritized using a multiobjective selection scheme that filters for selective and highly active compounds based on orthogonal methods grounded in computational chemistry and machine learning. Eighteen high-ranking compounds were synthesized and experimentally tested. The top-ranked compound has 10000-fold improved selectivity versus AurA, nanomolar cellular activity, and is highly selective in a kinase panel. This was achieved in a single round of automated in silico optimization, highlighting the power of recent advances in computer-aided drug design to automate design and selection processes.

Profiling Prediction of Kinase Inhibitors: Toward the Virtual Assay.

Kinome-wide screening would have the advantage of providing structure-activity relationships against hundreds of targets simultaneously. Here, we report the generation of ligand-based activity prediction models for over 280 kinases by employing Machine Learning methods on an extensive data set of proprietary bioactivity data combined with open data. High quality (AUC > 0.7) was achieved for ∼200 kinases by (1) combining open with proprietary data, (2) choosing Random Forest over alternative tested Machine Learning methods, and (3) balancing the training data sets. Tests on left-out and external data indicate a high value for virtual screening projects. Importantly, the derived models are evenly distributed across the kinome tree, allowing reliable profiling prediction for all kinase branches. The prediction quality was further improved by employing experimental bioactivity fingerprints of a small kinase subset. Overall, the generated models can support various hit identification tasks, including virtual screening, compound repurposing, and the detection of potential off-targets.

Unusual substrate specificity of the peptidoglycan MurE ligase from Erysipelothrix rhusiopathiae.

Erysipelothrix rhusiopathiae is a Gram-positive bacterium pathogenic to many species of birds and mammals, including humans. The main feature of its peptidoglycan is the presence of l-alanine at position 3 of the peptide stem. In the present work, we cloned the murE gene from E. rhusiopathiae and purified the corresponding protein as His6-tagged form. Enzymatic assays showed that E. rhusiopathiae MurE was indeed an l-alanine-adding enzyme. Surprisingly, it was also able, although to a lesser extent, to add meso-diaminopimelic acid, the amino acid found at position 3 in many Gram-negative bacteria, Bacilli and Mycobacteria. Sequence alignment of MurE enzymes from E. rhusiopathiae and Escherichia coli revealed that the DNPR motif that is characteristic of meso-diaminopimelate-adding enzymes was replaced by HDNR. The role of the latter motif in the interaction with l-alanine and meso-diaminopimelic acid was demonstrated by site-directed mutagenesis experiments and the construction of a homology model. The overexpression of the E. rhusiopathiae murE gene in E. coli resulted in the incorporation of l-alanine at position 3 of the peptide part of peptidoglycan.