Can we really do computer-aided drug design?

In this article, we discuss what we mean by ‘design’ and contrast this with the application of computational methods in drug discovery. We suggest that the predictivity of the computational models currently applied in drug discovery is not yet sufficient to permit a true design paradigm, as demonstrated by the large number of compounds that must currently be synthesised and tested to identify a successful drug. However, despite the uncertainties in predictions, computational methods have enormous potential value in narrowing the range of compounds to consider, by eliminating those that have negligible chance of being a successful drug, while focussing efforts on chemistries with the best likelihood of success. Applied appropriately, computational approaches can support decision-makers in achieving multi-parameter optimisation to guide the selection and design of compounds with the best chance of achieving an appropriate balance of properties for a drug discovery project’s objectives. Finally, we consider some approaches that may contribute over the next 25 years to improve the accuracy and transferability of computational models in drug discovery and move towards a genuine design process.     

Prediction of intestinal absorption and blood-brain barrier penetration by computational methods

This review surveys the computational methods that have been developed with the aim of identifying drug candidates likely to fail later on the road to market. The specifications for such computational methods are outlined, including factors such as speed, interpretability, robustness and accuracy. Then, computational filters aimed at predicting "drug-likeness" in a general sense are discussed before methods for the prediction of more specific properties--intestinal absorption and blood-brain barrier penetration--are reviewed. Directions for future research are discussed and, in concluding, the impact of these methods on the drug discovery process, both now and in the future, is briefly considered.     

Modeling & Informatics at Vertex Pharmaceuticals Incorporated: our philosophy for sustained impact

Molecular modelers and informaticians have the unique opportunity to integrate cross-functional data using a myriad of tools, methods and visuals to generate information. Using their drug discovery expertise, information is transformed to knowledge that impacts drug discovery. These insights are often times formulated locally and then applied more broadly, which influence the discovery of new medicines. This is particularly true in an organization where the members are exposed to projects throughout an organization, such as in the case of the global Modeling & Informatics group at Vertex Pharmaceuticals. From its inception, Vertex has been a leader in the development and use of computational methods for drug discovery. In this paper, we describe the Modeling & Informatics group at Vertex and the underlying philosophy, which has driven this team to sustain impact on the discovery of first-in-class transformative medicines.     

Computer-aided drug discovery research at a global contract research organization

Computer-aided drug discovery started at Albany Molecular Research, Inc in 1997. Over nearly 20 years the role of cheminformatics and computational chemistry has grown throughout the pharmaceutical industry and at AMRI. This paper will describe the infrastructure and roles of CADD throughout drug discovery and some of the lessons learned regarding the success of several methods. Various contributions provided by computational chemistry and cheminformatics in chemical library design, hit triage, hit-to-lead and lead optimization are discussed. Some frequently used computational chemistry techniques are described. The ways in which they may contribute to discovery projects are presented based on a few examples from recent publications.     

Machine Learning Methods in Drug Discovery

The advancements of information technology and related processing techniques have created a fertile base for progress in many scientific fields and industries. In the fields of drug discovery and development, machine learning techniques have been used for the development of novel drug candidates. The methods for designing drug targets and novel drug discovery now routinely combine machine learning and deep learning algorithms to enhance the efficiency, efficacy, and quality of developed outputs. The generation and incorporation of big data, through technologies such as high-throughput screening and high through-put computational analysis of databases used for both lead and target discovery, has increased the reliability of the machine learning and deep learning incorporated techniques. The use of these virtual screening and encompassing online information has also been highlighted in developing lead synthesis pathways. In this review, machine learning and deep learning algorithms utilized in drug discovery and associated techniques will be discussed. The applications that produce promising results and methods will be reviewed.     

Modern drug discovery technologies: opportunities and challenges in lead discovery

The identification of promising hits and the generation of high quality leads are crucial steps in the early stages of drug discovery projects. The definition and assessment of both chemical and biological space have revitalized the screening process model and emphasized the importance of exploring the intrinsic complementary nature of classical and modern methods in drug research. In this context, the widespread use of combinatorial chemistry and sophisticated screening methods for the discovery of lead compounds has created a large demand for small organic molecules that act on specific drug targets. Modern drug discovery involves the employment of a wide variety of technologies and expertise in multidisciplinary research teams. The synergistic effects between experimental and computational approaches on the selection and optimization of bioactive compounds emphasize the importance of the integration of advanced technologies in drug discovery programs. These technologies (VS, HTS, SBDD, LBDD, QSAR, and so on) are complementary in the sense that they have mutual goals, thereby the combination of both empirical and in silico efforts is feasible at many different levels of lead optimization and new chemical entity (NCE) discovery. This paper provides a brief perspective on the evolution and use of key drug design technologies, highlighting opportunities and challenges.     

Molecular simulation methods in drug discovery: a prospective outlook

Over the last decades, molecular simulations have spread through the drug discovery arena. This trend is expected to continue in the foreseeable future thanks to increased performance and the positive impact they can exert on productivity. In this article we highlight three aspects of molecular modelling for which we expect significant improvements over the next 25 years. Increased computational resources, faster algorithms and novel methods to sample rare events will provide a better handle on target flexibility and its relation with ligand binding. More accurate target druggability predictions will improve the success, but also broaden the scope of target-based drug discovery strategies. Finally, the use of higher levels of theory will increase the accuracy of protein–ligand binding affinity predictions, resulting in better hit identification success rates as well as more efficient lead optimization processes.     

Computer-aided drug design platform using PyMOL

The understanding and optimization of protein-ligand interactions are instrumental to medicinal chemists investigating potential drug candidates. Over the past couple of decades, many powerful standalone tools for computer-aided drug discovery have been developed in academia providing insight into protein-ligand interactions. As programs are developed by various research groups, a consistent user-friendly graphical working environment combining computational techniques such as docking, scoring, molecular dynamics simulations, and free energy calculations is needed. Utilizing PyMOL we have developed such a graphical user interface incorporating individual academic packages designed for protein preparation (AMBER package and Reduce), molecular mechanics applications (AMBER package), and docking and scoring (AutoDock Vina and SLIDE). In addition to amassing several computational tools under one interface, the computational platform also provides a user-friendly combination of different programs. For example, utilizing a molecular dynamics (MD) simulation performed with AMBER as input for ensemble docking with AutoDock Vina. The overarching goal of this work was to provide a computational platform that facilitates medicinal chemists, many who are not experts in computational methodologies, to utilize several common computational techniques germane to drug discovery. Furthermore, our software is open source and is aimed to initiate collaborative efforts among computational researchers to combine other open source computational methods under a single, easily understandable graphical user interface.     

Mini-review: computational structure-based design of inhibitors that target protein surfaces

Finding drugs that inhibit protein-protein interactions is usually difficult. While computer-aided design is used widely to facilitate the drug discovery process for protein targets with well-defined binding pockets, its application to the design of inhibitors targeting a protein surface is very limited. In this mini-review we address two aspects of this issue: firstly, we overview the current state of design methodology for inhibitors specifically targeting protein surfaces, and secondly, we briefly outline recent advances in computational methods for structure-based drug design. These methods are closely related to protein docking and protein recognition, the difference being that in ligand design, ligands are built on a fragment-by-fragment basis. A novel scheme of computational combinatorial ligand design developed for the design of inhibitors that interfere with protein-protein interaction is described in detail. Current applications and limitations of this methodology, as well as its future prospects, are discussed.     

Unmodified methodologies in target discovery for small molecule drugs: A rising star

Target discovery, involving target identification and validation, is the prerequisite for drug discovery and screening. Novel methodologies and technologies for the precise discovery and confirmation of drug targets are powerful tools in understanding the disease, looking for a drug and elucidating the mechanism of drug treatment. Among the common target identification and confirmation methods, the modified method is time-consuming and laborious, which may reduce or change the activity of natural products. The unmodified methods developed in recent years without chemical modification have gradually become an important means of studying drug targets. A wide range of unmodified approaches have been reported, introducing and analyzing the recent emerging methodologies and technologies. This review highlights the advantages and limitations of these methods for the application of drug target discovery and presents an overview of their contributions to the target discovery of small molecule drugs. The application and future development trends of methodologies in target discovery are also prospected to provide a reference for drug target research.     

Computer-aided drug design at Boehringer Ingelheim

Computer-Aided Drug Design (CADD) is an integral part of the drug discovery endeavor at Boehringer Ingelheim (BI). CADD contributes to the evaluation of new therapeutic concepts, identifies small molecule starting points for drug discovery, and develops strategies for optimizing hit and lead compounds. The CADD scientists at BI benefit from the global use and development of both software platforms and computational services. A number of computational techniques developed in-house have significantly changed the way early drug discovery is carried out at BI. In particular, virtual screening in vast chemical spaces, which can be accessed by combinatorial chemistry, has added a new option for the identification of hits in many projects. Recently, a new framework has been implemented allowing fast, interactive predictions of relevant on and off target endpoints and other optimization parameters. In addition to the introduction of this new framework at BI, CADD has been focusing on the enablement of medicinal chemists to independently perform an increasing amount of molecular modeling and design work. This is made possible through the deployment of MOE as a global modeling platform, allowing computational and medicinal chemists to freely share ideas and modeling results. Furthermore, a central communication layer called the computational chemistry framework provides broad access to predictive models and other computational services.     

High-throughput screening of enzyme inhibition using an inhibitor gradient generated in a microchannel

A new rapid microfluidic method for measuring enzyme inhibition is presented. The assay relies upon the creation of a uniform concentration of substrate and a microfluidically generated concentration gradient of inhibitor using a single microchannel and a single initial inhibitor concentration. The IC(50) values of two enzyme inhibitors were determined using the new technique and validated using a conventional microtiter plate assay. Using both experimental and computational simulation techniques, the assay was shown to be sensitive to inhibitor potency and the distribution of inhibitor in the system. The method has the potential to be more accurate than conventional methods because of the comparatively large amount of data that may be collected. Recommendations for use of the assay are provided, including its use for high-throughput screening in drug discovery and general use in measurement of enzyme inhibition.     

Hydra: a self regenerating high performance computing grid for drug discovery

Computer aided drug design is progressing and playing an increasingly important role in drug discovery. Computational methods are being used to evaluate larger and larger numbers of real and virtual compounds. New methods based on molecular simulations that take protein and ligand flexibility into account also contribute to an ever increasing need for compute time. Computational grids are therefore becoming a critically important tool for modern drug discovery, but can be expensive to deploy and maintain. Here, we describe the low cost implementation of a 165 node, computational grid at Anadys Pharmaceuticals. The grid makes use of the excess computing capacity of desktop computers deployed throughout the company and of outdated desktop computers which populate a central computing grid. The performance of the grid grows automatically with the size of the company and with advances in computer technology. To ensure the uniformity of the nodes in the grid, all computers are running the Linux operating system. The desktop computers run Linux inside MS Windows using coLinux as virtualization software. HYDRA has been used to optimize computational models, for virtual screening and for lead optimization.     

The need for scientific software engineering in the pharmaceutical industry

Scientific software engineering is a distinct discipline from both computational chemistry project support and research informatics. A scientific software engineer not only has a deep understanding of the science of drug discovery but also the desire, skills and time to apply good software engineering practices. A good team of scientific software engineers can create a software foundation that is maintainable, validated and robust. If done correctly, this foundation enable the organization to investigate new and novel computational ideas with a very high level of efficiency.     

Current trends in virtual high throughput screening using ligand-based and structure-based methods

High throughput in silico methods have offered the tantalizing potential to drastically accelerate the drug discovery process. Yet despite significant efforts expended by academia, national labs and industry over the years, many of these methods have not lived up to their initial promise of reducing the time and costs associated with the drug discovery enterprise, a process that can typically take over a decade and cost hundreds of millions of dollars from conception to final approval and marketing of a drug. Nevertheless structure-based modeling has become a mainstay of computational biology and medicinal chemistry, helping to leverage our knowledge of the biological target and the chemistry of protein-ligand interactions. While ligand-based methods utilize the chemistry of molecules that are known to bind to the biological target, structure-based drug design methods rely on knowledge of the three-dimensional structure of the target, as obtained through crystallographic, spectroscopic or bioinformatics techniques. Here we review recent developments in the methodology and applications of structure-based and ligand-based methods and target-based chemogenomics in Virtual High Throughput Screening (VHTS), highlighting some case studies of recent applications, as well as current research in further development of these methods. The limitations of these approaches will also be discussed, to give the reader an indication of what might be expected in years to come.     

Integrating virtual screening and combinatorial chemistry for accelerated drug discovery

Virtual screening is increasingly being used in drug discovery programs with a growing number of successful applications. Experimental methodologies developed to speed up the drug discovery processes include high-throughput screening and combinatorial chemistry. The complementarities between computational and experimental screenings have been recognized and reviewed in the literature. Computational methods have also been used in the combinatorial chemistry field, in particular in library design. However, the integration of computational and combinatorial chemistry screenings has been attempted only recently. Combinatorial libraries (experimental or virtual) represent a notable source of chemically related compounds. Advances in combinatorial chemistry and deconvolution strategies, have enabled the rapid exploration of novel and dense regions in the chemical space. The present review is focused on the integration of virtual and experimental screening of combinatorial libraries. Applications of virtual screening to discover novel anticancer agents and our ongoing efforts towards the integration of virtual screening and combinatorial chemistry are also discussed.     

Rational design of an enzyme mutant for anti-cocaine therapeutics

(-)-Cocaine is a widely abused drug and there is no available anti-cocaine therapeutic. The disastrous medical and social consequences of cocaine addiction have made the development of an effective pharmacological treatment a high priority. An ideal anti-cocaine medication would be to accelerate (-)-cocaine metabolism producing biologically inactive metabolites. The main metabolic pathway of cocaine in body is the hydrolysis at its benzoyl ester group. Reviewed in this article is the state-of-the-art computational design of high-activity mutants of human butyrylcholinesterase (BChE) against (-)-cocaine. The computational design of BChE mutants have been based on not only the structure of the enzyme, but also the detailed catalytic mechanisms for BChE-catalyzed hydrolysis of (-)-cocaine and (+)-cocaine. Computational studies of the detailed catalytic mechanisms and the structure-and-mechanism-based computational design have been carried out through the combined use of a variety of state-of-the-art techniques of molecular modeling. By using the computational insights into the catalytic mechanisms, a recently developed unique computational design strategy based on the simulation of the rate-determining transition state has been employed to design high-activity mutants of human BChE for hydrolysis of (-)-cocaine, leading to the exciting discovery of BChE mutants with a considerably improved catalytic efficiency against (-)-cocaine. One of the discovered BChE mutants (i.e., A199S/S287G/A328W/Y332G) has a approximately 456-fold improved catalytic efficiency against (-)-cocaine. The encouraging outcome of the computational design and discovery effort demonstrates that the unique computational design approach based on the transition-state simulation is promising for rational enzyme redesign and drug discovery.