A neural network based virtual screening of cytochrome P450 3A4 inhibitors

A virtual screening test to identify potential CP450 3A4 inhibitors has been developed. Molecular structures of inhibitors and non-inhibitors available in the Genetest database were represented using 2D Unity fingerprints and a feedforward neural network was trained to classify molecules regarding their inhibitory activity. Validation tests revealed that our neural net recognizes at least 89% of 3A4 inhibitors and suggest using this methodology in our virtual screening protocol.     

High-throughput ligand screening via preclustering and evolved neural networks

The pathway for novel lead drug discovery has many major deficiencies, the most significant of which is the immense size of small molecule diversity space. Methods that increase the search efficiency and/or reduce the size of the search space, increase the rate at which useful lead compounds are identified. Artificial neural networks optimized via evolutionary computation provide a cost and time-effective solution to this problem. Here, we present results that suggest preclustering of small molecules prior to neural network optimization is useful for generating models of quantitative structure-activity relationships for a set of HIV inhibitors. Using these methods, it is possible to prescreen compounds to separate active from inactive compounds or even actives and mildly active compounds from inactive compounds with high predictive accuracy while simultaneously reducing the feature space. It is also possible to identify "human interpretable" features from the best models that can be used for proposal and synthesis of new compounds in order to optimize potency and specificity.     

Second generation of BACE-1 inhibitors part 3: Towards non hydroxyethylamine transition state mimetics

Our first generation of hydroxyethylamine BACE-1 inhibitors proved unlikely to provide molecules that would lower amyloid in an animal model at low oral doses. This observation led us to the discovery of a second generation of inhibitors having nanomolar activity in a cell-based assay and with the potential for improved pharmacokinetic profiles. In this Letter, we describe our successful strategy for the optimization of oral bioavailability and also give insights into the design of compounds with the potential for improved brain penetration.     

A Multilayer Network Approach for Guiding Drug Repositioning in Neglected Diseases

Drug development for neglected diseases has been historically hampered due to lack of market incentives. The advent of public domain resources containing chemical information from high throughput screenings is changing the landscape of drug discovery for these diseases. In this work we took advantage of data from extensively studied organisms like human, mouse, E. coli and yeast, among others, to develop a novel integrative network model to prioritize and identify candidate drug targets in neglected pathogen proteomes, and bioactive drug-like molecules. We modeled genomic (proteins) and chemical (bioactive compounds) data as a multilayer weighted network graph that takes advantage of bioactivity data across 221 species, chemical similarities between 1.7 10⁵ compounds and several functional relations among 1.67 10⁵ proteins. These relations comprised orthology, sharing of protein domains, and shared participation in defined biochemical pathways. We showcase the application of this network graph to the problem of prioritization of new candidate targets, based on the information available in the graph for known compound-target associations. We validated this strategy by performing a cross validation procedure for known mouse and Trypanosoma cruzi targets and showed that our approach outperforms classic alignment-based approaches. Moreover, our model provides additional flexibility as two different network definitions could be considered, finding in both cases qualitatively different but sensible candidate targets. We also showcase the application of the network to suggest targets for orphan compounds that are active against Plasmodium falciparum in high-throughput screens. In this case our approach provided a reduced prioritization list of target proteins for the query molecules and showed the ability to propose new testable hypotheses for each compound. Moreover, we found that some predictions highlighted by our network model were supported by independent experimental validations as found post-facto in the literature.     

Prediction of inhibitor binding free energies by quantum neural networks. Nucleoside analogues binding to trypanosomal nucleoside hydrolase

A computational method has been developed to predict inhibitor binding energy for untested inhibitor molecules. A neural network is trained from the electrostatic potential surfaces of known inhibitors and their binding energies. The algorithm is then able to predict, with high accuracy, the binding energy of unknown inhibitors. IU-nucleoside hydrolase from Crithidia fasciculata and the inhibitor molecules described previously [Miles, R. W. Tyler, P. C. Evans, G. Furneaux R. H., Parkin, D. W., and Schramm, V. L. (1999) Biochemistry 38, xxxx-xxxx] are used as the test system. Discrete points on the molecular electrostatic potential surface of inhibitor molecules are input to neural networks to identify the quantum mechanical features that contribute to binding. Feed-forward neural networks with back-propagation of error are trained to recognize the quantum mechanical electrostatic potential and geometry at the entire van der Waals surface of a group of training molecules and to predict the strength of interactions between the enzyme and novel inhibitors. The binding energies of unknown inhibitors were predicted, followed by experimental determination of K(i)() values. Predictions of K(i)() values using this theory are compared to other methods and are more robust in estimating inhibitory strength. The average deviation in estimating K(i)() values for 18 unknown inhibitor molecules, with 21 training molecules, is a factor of 5 x K(i)() over a range of 660 000 in K(i)() values for all molecules. The a posteriori accuracy of the predictions suggests the method will be effective as a guide for experimental inhibitor design.     

Design,synthesis, and enzymatic characterization of quinazoline-based CYP1A2 inhibitors

Cytochrome P450 isozyme 1A2 (CYP1A2) is one main xenobiotic metabolizing enzyme in humans. It has been associated with the bioactivation of procarcinogens, including 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone (NNK), a tobacco specific and potent pulmonary carcinogen. This work describes the computational design and in-silico screening of potential CYP1A2 inhibitors, their chemical synthesis, and enzymatic characterization with the ultimate aim of assessing their potential as cancer chemopreventive agents. To achieve this, a combined classifiers model was used to screen a library of quinazoline-based molecules against known CYP1A2 inhibitors, non-inhibitors, and substrates to predict which quinazoline candidates had a better probability as an inhibitor. Compounds with high probability of CYP1A2 inhibition were further computationally evaluated via Glide docking. Candidates predicted to have selectivity and high binding affinity for CYP1A2 were synthesized and assayed for their enzymatic inhibition of CYP1A2, leading to the discovery of novel and potent quinazoline-based CYP1A2 inhibitors.     

Discovery of hydroxamic acid analogs as dual inhibitors of phosphodiesterase-1 and -5

HTS and the following synthesis of a series of the compounds led us to the discovery of hydroxamic acid analogs as potent dual inhibitors of phosphodiesterase (PDE)-1 and 5. These compounds have highly related structure and deviation of the structure usually resulted in reduced potency. This result can be used to design other molecules that may be utilized for the therapy of cardiovascular symptoms that relates to cGMP level.     

Drug screening for influenza neuraminidase inhibitors

Neuraminidase (NA) is one of the most important targets to screen the drugs of anti-influenza virus A and B. After virtual screening approaches were applied to a compound database which possesses more than 10000 compound structures, 160 compounds were selected for bioactivity assay, then a High Throughput Screening (HTS) model established for influenza virus NA inhibitors was applied to detect these compounds. Finally, three compounds among them displayed higher inhibitory activities, the range of their IC50 was from 0.1 μmol/L to 3μmol/L. Their structural scaffolds are novel and different from those of NA inhibitors approved for influenza treatment, and will be useful for the design and research of new NA inhibitors. The resuit indicated that the combination of virtual screening with HTS was very significant to drug screening and drug discovery.     

Discovery of tricyclic 5,6-dihydro-1H-pyridin-2-ones as novel, potent, and orally bioavailable inhibitors of HCV NS5B polymerase

A novel series of non-nucleoside small molecules containing a tricyclic dihydropyridinone structural motif was identified as potent HCV NS5B polymerase inhibitors. Driven by structure-based design and building on our previous efforts in related series of molecules, we undertook extensive SAR studies, in which we identified a number of metabolically stable and very potent compounds in genotype 1a and 1b replicon assays. This work culminated in the discovery of several inhibitors, which combined potent in vitro antiviral activity against both 1a and 1b genotypes, metabolic stability, good oral bioavailability, and high C₁₂ (PO)/EC₅₀ ratios.     

State-based discovery: a multidimensional screen for small-molecule modulators of EGF signaling

As an alternative to conventional, target-oriented drug discovery, we report a strategy that identifies compounds on the basis of the state that they induce in a signaling network. Immortalized human cells are grown in microtiter plates and treated with compounds from a small-molecule library. The target network is then activated and lysates derived from each sample are arrayed onto glass-supported nitrocellulose pads. By probing these microarrays with antibodies that report on the abundance or phosphorylation state of selected proteins, a global picture of the target network is obtained. As proof of concept, we screened 84 kinase and phosphatase inhibitors for their ability to induce different states in the ErbB signaling network. We observed functional connections between proteins that match our understanding of ErbB signaling, indicating that state-based screens can be used to define the topology of signaling networks. Additionally, compounds sort according to the multidimensional phenotypes they induce, suggesting that state-based screens may inform efforts to identify the targets of biologically active small molecules.     

When Inhibitors Do Not Inhibit: Critical Evaluation of Rational Drug Design Targeting Chorismate Mutase from Mycobacterium tuberculosis

Tuberculosis (TB) is a devastating disease that claims millions of lives every year. Hindered access or non‐compliance to medication, especially in developing countries, led to drug resistance, further aggravating the situation. With current standard therapies in use for over 50 years and only few new candidates in clinical trials, there is an urgent call for new TB drugs. A powerful tool for the development of new medication is structure‐guided design, combined with virtual screening or docking studies. Here, we report the results of a drug‐design project, which we based on a publication that claimed the structure‐guided discovery of several promising and highly active inhibitors targeting the secreted chorismate mutase (*MtCM) from Mycobacterium tuberculosis. We set out to further improve on these compounds and synthesized a series of new derivatives. Thorough evaluation of these molecules in enzymatic assays revealed, to our dismay, that neither the claimed lead compounds, nor any of the synthesized derivatives, show any inhibitory effects against *MtCM.     

Compounds for selective translational inhibition

Since many human diseases are caused by the unwelcome production of harmful proteins, compounds that selectively suppress protein synthesis should provide a unique path for drug development, expanding the druggable proteome. Although surveying the RNA/amino acid contexts that are preferentially affected by translation inhibitors has presented an analytic hurdle, the application of a technique termed ribosome profiling overcomes this problem. Indeed, this technique uncovers the selectivity of translation repression by small molecules such as chloramphenicol, macrolides, PF846, and rocaglates. The molecular understanding of how the compounds inspire context selectivity, despite their targeting to general translation machinery, facilitates rational drug design and discovery for therapeutic purposes.     

Design, synthesis and biological evaluation of peptidyl-vinylaminophosphonates as novel cysteine protease inhibitors

We report herein, design and synthesis of vinylaminophosphonates, a novel class of compounds as possible cysteine protease inhibitors. The synthesis of vinylaminophosphonates has been accomplished employing Tsuji-Trost reaction as a key step. The synthesized compounds were assayed against papain, a model cysteine protease and some of our synthesized compounds showed IC(50) values in the range of 30-54 μM thereby suggesting that these chemical entities thus could constitute an interesting template for the design of potential novel protease inhibitors.     

A dual fingerprint-based metric for the design of focused compound libraries and analogs

A computational metric is introduced for the design of combinatorial libraries focused on small molecules with specific activity (e.g., enzyme inhibitors). The method follows a product-based design strategy and uses combinations of two binary molecular fingerprints to create chemical diversity around selected compounds and/or core structures. In the first step, compounds are sampled that are distinct from template molecules but likely to share similar biological activity. In the second step, designed compounds are accepted if they are not too similar to each other, as assessed by calculation of fingerprint overlap. Thus, it is possible to balance molecular "similarity" and "diversity" and control the degree of chemical diversity created in the vicinity of selected template molecules. In essence, the method aims to generate diverse arrays of compounds with a high probability of having activity similar to starting molecule(s) and is therefore well suited for the design of target-focused libraries or series of analogs. As an example, the method is applied to focus libraries on known protein kinase inhibitors.     

Production of random DNA oligomers for scalable DNA computing

While remarkably complex networks of connected DNA molecules can form from a relatively small number of distinct oligomer strands, a large computational space created by DNA reactions would ultimately require the use of many distinct DNA strands. The automatic synthesis of this many distinct strands is economically prohibitive. We present here a new approach to producing distinct DNA oligomers based on the polymerase chain reaction (PCR) amplification of a few random template sequences. As an example, we designed a DNA template sequence consisting of a 50-mer random DNA segment flanked by two 20-mer invariant primer sequences. Amplification of a dilute sample containing about 30 different template molecules allows us to obtain around 10¹¹ copies of these molecules and their complements. We demonstrate the use of these amplicons to implement some of the vector operations that will be required in a DNA implementation of an analog neural network.