Enabling future drug discovery by de novo design

Revolutionizing adjuvant development: harnessing AI for next-generation cancer vaccines

With the COVID-19 pandemic, the importance of vaccines has been widely recognized and has led to increased research and development efforts. Vaccines also play a crucial role in cancer treatment by activating the immune system to target and destroy cancer cells. However, enhancing the efficacy of cancer vaccines remains a challenge. Adjuvants, which enhance the immune response to antigens and improve vaccine effectiveness, have faced limitations in recent years, resulting in few novel adjuvants being identified. The advancement of artificial intelligence (AI) technology in drug development has provided a foundation for adjuvant screening and application, leading to a diversification of adjuvants. This article reviews the significant role of tumor vaccines in basic research and clinical treatment and explores the use of AI technology to screen novel adjuvants from databases. The findings of this review offer valuable insights for the development of new adjuvants for next-generation vaccines.

De Novo Design of Inhibitors of DNA Methyltransferase 1: A Critical Comparison of Ligand- and Structure-Based Approaches

Designing and developing inhibitors against the epigenetic target DNA methyltransferase (DNMT) is an attractive strategy in epigenetic drug discovery. DNMT1 is one of the epigenetic enzymes with significant clinical relevance. Structure-based de novo design is a drug discovery strategy that was used in combination with similarity searching to identify a novel DNMT inhibitor with a novel chemical scaffold and warrants further exploration. This study aimed to continue exploring the potential of de novo design to build epigenetic-focused libraries targeted toward DNMT1. Herein, we report the results of an in-depth and critical comparison of ligand- and structure-based de novo design of screening libraries focused on DNMT1. The newly designed chemical libraries focused on DNMT1 are freely available on GitHub.

Artificial design of organic emitters via a genetic algorithm enhanced by a deep neural network

The design of molecules requires multi-objective optimizations in high-dimensional chemical space with often conflicting target properties. To navigate this space, classical workflows rely on the domain knowledge and creativity of human experts, which can be the bottleneck in high-throughput approaches. Herein, we present an artificial molecular design workflow relying on a genetic algorithm and a deep neural network to find a new family of organic emitters with inverted singlet-triplet gaps and appreciable fluorescence rates. We combine high-throughput virtual screening and inverse design infused with domain knowledge and artificial intelligence to accelerate molecular generation significantly. This enabled us to explore more than 800 000 potential emitter molecules and find more than 10 000 candidates estimated to have inverted singlet-triplet gaps (INVEST) and appreciable fluorescence rates, many of which likely emit blue light. This class of molecules has the potential to realize a new generation of organic light-emitting diodes.

Efficient Computational Framework for Target-Specific Active Peptide Discovery: A Case Study on IL-17C Targeting Cyclic Peptides

The development of potentially active peptides for specific targets is critical for the modern pharmaceutical industry's growth. In this study, we present an efficient computational framework for the discovery of active peptides targeting a specific pharmacological target, which combines a conditional variational autoencoder (CVAE) and a classifier named TCPP based on the Transformer and convolutional neural network. In our example scenario, we constructed an active cyclic peptide library targeting interleukin-17C (IL-17C) through a library-based in vitro selection strategy. The CVAE model is trained on the preprocessed peptide data sets to generate potentially active peptides and the TCPP further screens the generated peptides. Ultimately, six candidate peptides predicted by the model were synthesized and assayed for their activity, and four of them exhibited promising binding affinity to IL-17C. Our study provides a one-stop-shop for target-specific active peptide discovery, which is expected to boost up the process of peptide drug development.

Fake it until you make it? Generative de novo design and virtual screening of synthesizable molecules

Computational techniques, including virtual screening, de novo design, and generative models, play an increasing role in expediting DMTA cycles for modern molecular discovery. However, computationally proposed molecules must be synthetically feasible for laboratory testing. In this perspective, we offer a succinct introduction to the subject, and showcase typical workflows to integrate synthesis planning, synthesizability scoring, and molecule generation. Finally, we address limitations and opportunities for future research.

Descriptor-Driven de Novo Design Algorithms for DOCK6 Using RDKit

Structure-based methods that employ principles of de novo design can be used to construct small organic molecules from scratch using pre-existing fragment libraries to sample chemical space and are an important class of computational algorithms for drug-lead discovery. Here, we present a powerful new design method for DOCK6 that employs a Descriptor-Driven De Novo strategy (termed D3N) in which user-defined cheminformatics descriptors (and their target ranges) are calculated at each layer of growth using the open-source toolkit RDKit. The objective is to tailor ligand growth toward desirable regions of chemical space. The approach was extensively validated through: (1) comparison of cheminformatics descriptors computed using the new DOCK6/RDKit interface versus the standard Python/RDKit installation, (2) examination of descriptor distributions generated using D3N growth under different conditions (target ranges and environments), and (3) construction of ligands with very tight (pinpoint) descriptor ranges using clinically relevant compounds as a reference. Our testing confirms that the new DOCK6/RDKit integration is robust, showcases how the new D3N routines can be used to direct sampling around user-defined chemical spaces, and highlights the utility of on-the-fly descriptor calculations for ligand design to important drug targets.

From theory to experiment: transformer-based generation enables rapid discovery of novel reactions

Deep learning methods, such as reaction prediction and retrosynthesis analysis, have demonstrated their significance in the chemical field. However, the de novo generation of novel reactions using artificial intelligence technology requires further exploration. Inspired by molecular generation, we proposed a novel task of reaction generation. Herein, Heck reactions were applied to train the transformer model, a state-of-art natural language process model, to generate 4717 reactions after sampling and processing. Then, 2253 novel Heck reactions were confirmed by organizing chemists to judge the generated reactions. More importantly, further organic synthesis experiments were performed to verify the accuracy and feasibility of representative reactions. The total process, from Heck reaction generation to experimental verification, required only 15 days, demonstrating that our model has well-learned reaction rules in-depth and can contribute to novel reaction discovery and chemical space exploration.

Parallel tempered genetic algorithm guided by deep neural networks for inverse molecular design

Inverse molecular design involves algorithms that sample molecules with specific target properties from a multitude of candidates and can be posed as an optimization problem. High-dimensional optimization tasks in the natural sciences are commonly tackled via population-based metaheuristic optimization algorithms such as evolutionary algorithms. However, often unavoidable expensive property evaluation can limit the widespread use of such approaches as the associated cost can become prohibitive. Herein, we present JANUS, a genetic algorithm inspired by parallel tempering. It propagates two populations, one for exploration and another for exploitation, improving optimization by reducing property evaluations. JANUS is augmented by a deep neural network that approximates molecular properties and relies on active learning for enhanced molecular sampling. It uses the SELFIES representation and the STONED algorithm for the efficient generation of structures, and outperforms other generative models in common inverse molecular design tasks achieving state-of-the-art target metrics across multiple benchmarks. As neither most of the benchmarks nor the structure generator in JANUS account for synthesizability, a significant fraction of the proposed molecules is synthetically infeasible demonstrating that this aspect needs to be considered when evaluating the performance of molecular generative models.

DeLA-Drug: A Deep Learning Algorithm for Automated Design of Druglike Analogues

In this paper, we present a deep learning algorithm for automated design of druglike analogues (DeLA-Drug), a recurrent neural network (RNN) model composed of two long short-term memory (LSTM) layers and conceived for data-driven generation of similar-to-bioactive compounds. DeLA-Drug captures the syntax of SMILES strings of more than 1 million compounds belonging to the ChEMBL28 database and, by employing a new strategy called sampling with substitutions (SWS), generates molecules starting from a single user-defined query compound. Remarkably, the algorithm preserves druglikeness and synthetic accessibility of the known bioactive compounds present in the ChEMBL28 repository. The absence of any time-demanding fine-tuning procedure enables DeLA-Drug to perform a fast generation of focused libraries for further high-throughput screening and makes it a suitable tool for performing de novo design even in low-data regimes. To provide a concrete idea of its applicability, DeLA-Drug was applied to the cannabinoid receptor subtype 2 (CB2R), a known target involved in different pathological conditions such as cancer and neurodegeneration. DeLA-Drug, available as a free web platform (http://www.ba.ic.cnr.it/softwareic/deladrugportal/), can help medicinal chemists interested in generating analogues of compounds already available in their laboratories and, for this reason, good candidates for an easy and low-cost synthesis.

Regulation of Nrf2 and NF-κB activities may contribute to the anti-inflammatory mechanism of xylopic acid

Xylopic acid (XA) is a kaurene diterpene which naturally exists in African plants such as Xylopia aethiopica. It has been established to exhibit acute and chronic anti-inflammatory activities from our earlier studies. This current work sets out to shed light on the potential molecular target(s) of xylopic acid. Selection of investigated targets (NF-κB, Nrf2 and PTP1B) was based on an unbiased approach, using the SPiDER in silico prediction tool, and a candidate approach, examining well-known anti-inflammatory targets. Reporter gene assays were used to test for altered NF-κB and Nrf2 activities in transfected HEK or CHO cells, respectively, and immunoblot and flow cytometric analyses examined protein expression of the Nrf2/NF-kB target genes HO-1 and VCAM-1 in HUVEC. An effect of XA on PTP1B activity assay was studied using an in vitro enzyme assay with recombinant human enzyme and pNPP as substrate as well as by looking at insulin receptor phosphorylation in HepG2 cells. XA at 30 µM significantly (p < 0.001) inhibited the NF-κB-dependent reporter gene expression and enhanced activation of Nrf2 in a concentration-dependent manner when compared to the control. XA also marginally increased HO-1 protein expression levels while expression of VCAM-1 was reduced to 70% in XA-treated endothelial cells. However, XA did not show any sign of inhibition of PTP1B or a related phosphatase. Our findings suggest that the anti-inflammatory mechanism of XA entails the inhibitory effect on NF-κB and an increased activity of Nrf2, accompanied by increased expression of HO-1 and reduced expression of VCAM-1.

Artificial Intelligence in Vaccine and Drug Design

Knowledge in the fields of biochemistry, structural biology, immunological principles, microbiology, and genomics has all increased dramatically in recent years. There has also been tremendous growth in the fields of data science, informatics, and artificial intelligence needed to handle this immense data flow. At the intersection of wet lab and data science is the field of bioinformatics, which seeks to apply computational tools to better understanding of the biological sciences. Like so many other areas of biology, bioinformatics has transformed immunology research leading to the discipline of immunoinformatics. Within this field, many new databases and computational tools have been created that increasingly drive immunology research, in many cases drawing upon artificial intelligence and machine learning to predict complex immune system behaviors, for example, prediction of B cell and T cell epitopes. In this book chapter, we provide an overview of computational tools and artificial intelligence being used for protein modeling, drug screening, vaccine design, and highlight how these tools are being used to transform approaches to pandemic countermeasure development, by reference to the current COVID-19 pandemic.

Individual and collective human intelligence in drug design: evaluating the search strategy

In recent years, individual and collective human intelligence, defined as the knowledge, skills, reasoning and intuition of individuals and groups, have been used in combination with computer algorithms to solve complex scientific problems. Such approach was successfully used in different research fields such as: structural biology, comparative genomics, macromolecular crystallography and RNA design. Herein we describe an attempt to use a similar approach in small-molecule drug discovery, specifically to drive search strategies of de novo drug design. This is assessed with a case study that consists of a series of public experiments in which participants had to explore the huge chemical space in silico to find predefined compounds by designing molecules and analyzing the score associate with them. Such a process may be seen as an instantaneous surrogate of the classical design-make-test cycles carried out by medicinal chemists during the drug discovery hit to lead phase but not hindered by long synthesis and testing times. We present first findings on (1) assessing human intelligence in chemical space exploration, (2) comparing individual and collective human intelligence performance in this task and (3) contrasting some human and artificial intelligence achievements in de novo drug design.

De novo molecular drug design benchmarking

De novo molecular design for drug discovery is a growing field. Deep neural networks (DNNs) are becoming more widespread in their use for machine learning models. As more DNN models are proposed for molecular design, benchmarking methods are crucial for the comparision and validation of these models. This review looks at recently proposed benchmarking methods Fréchet ChemNet Distance, GuacaMol and Molecular Sets (MOSES), and provides a commentary on their future potential applications in de novo molecular drug design and possible next steps for further validation of these benchmarking methods.

Artificial intelligence to deep learning: machine intelligence approach for drug discovery

Drug designing and development is an important area of research for pharmaceutical companies and chemical scientists. However, low efficacy, off-target delivery, time consumption, and high cost impose a hurdle and challenges that impact drug design and discovery. Further, complex and big data from genomics, proteomics, microarray data, and clinical trials also impose an obstacle in the drug discovery pipeline. Artificial intelligence and machine learning technology play a crucial role in drug discovery and development. In other words, artificial neural networks and deep learning algorithms have modernized the area. Machine learning and deep learning algorithms have been implemented in several drug discovery processes such as peptide synthesis, structure-based virtual screening, ligand-based virtual screening, toxicity prediction, drug monitoring and release, pharmacophore modeling, quantitative structure-activity relationship, drug repositioning, polypharmacology, and physiochemical activity. Evidence from the past strengthens the implementation of artificial intelligence and deep learning in this field. Moreover, novel data mining, curation, and management techniques provided critical support to recently developed modeling algorithms. In summary, artificial intelligence and deep learning advancements provide an excellent opportunity for rational drug design and discovery process, which will eventually impact mankind. The primary concern associated with drug design and development is time consumption and production cost. Further, inefficiency, inaccurate target delivery, and inappropriate dosage are other hurdles that inhibit the process of drug delivery and development. With advancements in technology, computer-aided drug design integrating artificial intelligence algorithms can eliminate the challenges and hurdles of traditional drug design and development. Artificial intelligence is referred to as superset comprising machine learning, whereas machine learning comprises supervised learning, unsupervised learning, and reinforcement learning. Further, deep learning, a subset of machine learning, has been extensively implemented in drug design and development. The artificial neural network, deep neural network, support vector machines, classification and regression, generative adversarial networks, symbolic learning, and meta-learning are examples of the algorithms applied to the drug design and discovery process. Artificial intelligence has been applied to different areas of drug design and development process, such as from peptide synthesis to molecule design, virtual screening to molecular docking, quantitative structure-activity relationship to drug repositioning, protein misfolding to protein-protein interactions, and molecular pathway identification to polypharmacology. Artificial intelligence principles have been applied to the classification of active and inactive, monitoring drug release, pre-clinical and clinical development, primary and secondary drug screening, biomarker development, pharmaceutical manufacturing, bioactivity identification and physiochemical properties, prediction of toxicity, and identification of mode of action.

Transmol: repurposing a language model for molecular generation

Recent advances in convolutional neural networks have inspired the application of deep learning to other disciplines. Even though image processing and natural language processing have turned out to be the most successful, there are many other domains that have also benefited; among them, life sciences in general and chemistry and drug design in particular. In concordance with this observation, from 2018 the scientific community has seen a surge of methodologies related to the generation of diverse molecular libraries using machine learning. However to date, attention mechanisms have not been employed for the problem of de novo molecular generation. Here we employ a variant of transformers, an architecture recently developed for natural language processing, for this purpose. Our results indicate that the adapted Transmol model is indeed applicable for the task of generating molecular libraries and leads to statistically significant increases in some of the core metrics of the MOSES benchmark. The presented model can be tuned to either input-guided or diversity-driven generation modes by applying a standard one-seed and a novel two-seed approach, respectively. Accordingly, the one-seed approach is best suited for the targeted generation of focused libraries composed of close analogues of the seed structure, while the two-seeds approach allows us to dive deeper into under-explored regions of the chemical space by attempting to generate the molecules that resemble both seeds. To gain more insights about the scope of the one-seed approach, we devised a new validation workflow that involves the recreation of known ligands for an important biological target vitamin D receptor. To further benefit the chemical community, the Transmol algorithm has been incorporated into our cheML.io web database of ML-generated molecules as a second generation on-demand methodology.

MolFinder: an evolutionary algorithm for the global optimization of molecular properties and the extensive exploration of chemical space using SMILES

Here, we introduce a new molecule optimization method, MolFinder, based on an efficient global optimization algorithm, the conformational space annealing algorithm, and the SMILES representation. MolFinder finds diverse molecules with desired properties efficiently without any training and a large molecular database. Compared with recently proposed reinforcement-learning-based molecule optimization algorithms, MolFinder consistently outperforms in terms of both the optimization of a given target property and the generation of a set of diverse and novel molecules. The efficiency of MolFinder demonstrates that combinatorial optimization using the SMILES representation is a promising approach for molecule optimization, which has not been well investigated despite its simplicity. We believe that our results shed light on new possibilities for advances in molecule optimization methods.

Scaffold-Constrained Molecular Generation

One of the major applications of generative models for drug discovery targets the lead-optimization phase. During the optimization of a lead series, it is common to have scaffold constraints imposed on the structure of the molecules designed. Without enforcing such constraints, the probability of generating molecules with the required scaffold is extremely low and hinders the practicality of generative models for de novo drug design. To tackle this issue, we introduce a new algorithm, named SAMOA (Scaffold Constrained Molecular Generation), to perform scaffold-constrained in silico molecular design. We build on the well-known SMILES-based Recurrent Neural Network (RNN) generative model, with a modified sampling procedure to achieve scaffold-constrained generation. We directly benefit from the associated reinforcement learning methods, allowing to design molecules optimized for different properties while exploring only the relevant chemical space. We showcase the method's ability to perform scaffold-constrained generation on various tasks: designing novel molecules around scaffolds extracted from SureChEMBL chemical series, generating novel active molecules on the Dopamine Receptor D2 (DRD2) target, and finally, designing predicted actives on the MMP-12 series, an industrial lead-optimization project.

Design of novel ROCK inhibitors using fragment-based de novo drug design approach

Rho-associated coiled-coil protein kinase (ROCK) is playing a vital role in the regulation of key cellular events and also responsible for causing several pathological conditions such as cancer, hypertension, Alzheimer's, cerebral vasospasm, and cardiac stroke. Therefore, it has attracted us to target ROCK protein as a potential therapeutic target for combating various diseases. Consequently, we investigated the active site of ROCK I protein and designed novel leads against the target using the de novo evolution drug design approach. Caffeic acid (an aglycone of acteoside) as a scaffold and fragments from 336 reported ROCK inhibitors were used for the design of novel leads. Multiple copy simultaneous search docking was used to identify the suitable fragments to be linked with the scaffold. Basic medicinal chemistry rules, coupled with structural insights generated by docking, led to the design of 7a, 8a, 9a, and 10a as potential ROCK I inhibitors. The designed leads showed better binding than the approved drug fasudil and also interacted with the key hinge region residue Met156 of ROCK I. Further, molecular dynamics (MD) simulation revealed that the protein-ligand complexes were stable and maintained the hydrogen bond with Met156 throughout the MD run. The promising in silico outcomes suggest that the designed compounds could be suitable anti-cancer leads that need to be synthesized and tested in various cancer cell lines. Graphical abstract.

CReM: chemically reasonable mutations framework for structure generation

Structure generators are widely used in de novo design studies and their performance substantially influences an outcome. Approaches based on the deep learning models and conventional atom-based approaches may result in invalid structures and fail to address their synthetic feasibility issues. On the other hand, conventional reaction-based approaches result in synthetically feasible compounds but novelty and diversity of generated compounds may be limited. Fragment-based approaches can provide both better novelty and diversity of generated compounds but the issue of synthetic complexity of generated structure was not explicitly addressed before. Here we developed a new framework of fragment-based structure generation that, by design, results in the chemically valid structures and provides flexible control over diversity, novelty, synthetic complexity and chemotypes of generated compounds. The framework was implemented as an open-source Python module and can be used to create custom workflows for the exploration of chemical space.

Reaction-Based Enumeration, Active Learning, and Free Energy Calculations To Rapidly Explore Synthetically Tractable Chemical Space and Optimize Potency of Cyclin-Dependent Kinase 2 Inhibitors

The hit-to-lead and lead optimization processes usually involve the design, synthesis, and profiling of thousands of analogs prior to clinical candidate nomination. A hit finding campaign may begin with a virtual screen that explores millions of compounds, if not more. However, this scale of computational profiling is not frequently performed in the hit-to-lead or lead optimization phases of drug discovery. This is likely due to the lack of appropriate computational tools to generate synthetically tractable lead-like compounds in silico, and a lack of computational methods to accurately profile compounds prospectively on a large scale. Recent advances in computational power and methods provide the ability to profile much larger libraries of ligands than previously possible. Herein, we report a new computational technique, referred to as "PathFinder", that uses retrosynthetic analysis followed by combinatorial synthesis to generate novel compounds in synthetically accessible chemical space. In this work, the integration of PathFinder-driven compound generation, cloud-based FEP simulations, and active learning are used to rapidly optimize R-groups, and generate new cores for inhibitors of cyclin-dependent kinase 2 (CDK2). Using this approach, we explored >300 000 ideas, performed >5000 FEP simulations, and identified >100 ligands with a predicted IC₅₀ < 100 nM, including four unique cores. To our knowledge, this is the largest set of FEP calculations disclosed in the literature to date. The rapid turnaround time, and scale of chemical exploration, suggests that this is a useful approach to accelerate the discovery of novel chemical matter in drug discovery campaigns.

Approaching Target Selectivity by De Novo Drug Design

Introduction: The development of drug candidates with a defined selectivity profile and a unique molecular structure is of fundamental interest for drug discovery. In contrast to the costly screening of large substance libraries, the targeted de novo design of a drug by using structural information of either the biological target and/or structure-activity relationship data of active modulators offers an efficient and intellectually appealing alternative. Areas covered: This review provides an overview on the different techniques of de novo drug design (ligand-based drug design, structure-based drug design, and fragment-based drug design) and highlights successful examples of this targeted approach toward selective modulators of therapeutically relevant targets. Expert opinion: De novo drug design has established itself as a very efficient method for the development of potent and selective modulators for a variety of different biological target classes. The ever-growing wealth of structural data on therapeutic targets will certainly further enhance the importance of de novo design for the drug discovery process in the future. However, a consistent use of the terminology of de novo drug design in the scientific literature should be sought.

GuacaMol: Benchmarking Models for de Novo Molecular Design

De novo design seeks to generate molecules with required property profiles by virtual design-make-test cycles. With the emergence of deep learning and neural generative models in many application areas, models for molecular design based on neural networks appeared recently and show promising results. However, the new models have not been profiled on consistent tasks, and comparative studies to well-established algorithms have only seldom been performed. To standardize the assessment of both classical and neural models for de novo molecular design, we propose an evaluation framework, GuacaMol, based on a suite of standardized benchmarks. The benchmark tasks encompass measuring the fidelity of the models to reproduce the property distribution of the training sets, the ability to generate novel molecules, the exploration and exploitation of chemical space, and a variety of single and multiobjective optimization tasks. The benchmarking open-source Python code and a leaderboard can be found on https://benevolent.ai/guacamol .

In Silico Target Prediction for Small Molecules

Drugs modulate disease states through their actions on targets in the body. Determining these targets aids the focused development of new treatments, and helps to better characterize those already employed. One means of accomplishing this is through the deployment of in silico methodologies, harnessing computational analytical and predictive power to produce educated hypotheses for experimental verification. Here, we provide an overview of the current state of the art, describe some of the well-established methods in detail, and reflect on how they, and emerging technologies promoting the incorporation of complex and heterogeneous data-sets, can be employed to improve our understanding of (poly)pharmacology.

Artificial Intelligence in Drug Design

Artificial Intelligence (AI) plays a pivotal role in drug discovery. In particular artificial neural networks such as deep neural networks or recurrent networks drive this area. Numerous applications in property or activity predictions like physicochemical and ADMET properties have recently appeared and underpin the strength of this technology in quantitative structure-property relationships (QSPR) or quantitative structure-activity relationships (QSAR). Artificial intelligence in de novo design drives the generation of meaningful new biologically active molecules towards desired properties. Several examples establish the strength of artificial intelligence in this field. Combination with synthesis planning and ease of synthesis is feasible and more and more automated drug discovery by computers is expected in the near future.

Molecular generative model based on conditional variational autoencoder for de novo molecular design

We propose a molecular generative model based on the conditional variational autoencoder for de novo molecular design. It is specialized to control multiple molecular properties simultaneously by imposing them on a latent space. As a proof of concept, we demonstrate that it can be used to generate drug-like molecules with five target properties. We were also able to adjust a single property without changing the others and to manipulate it beyond the range of the dataset.

Opportunities and obstacles for deep learning in biology and medicine

Deep learning describes a class of machine learning algorithms that are capable of combining raw inputs into layers of intermediate features. These algorithms have recently shown impressive results across a variety of domains. Biology and medicine are data-rich disciplines, but the data are complex and often ill-understood. Hence, deep learning techniques may be particularly well suited to solve problems of these fields. We examine applications of deep learning to a variety of biomedical problems-patient classification, fundamental biological processes and treatment of patients-and discuss whether deep learning will be able to transform these tasks or if the biomedical sphere poses unique challenges. Following from an extensive literature review, we find that deep learning has yet to revolutionize biomedicine or definitively resolve any of the most pressing challenges in the field, but promising advances have been made on the prior state of the art. Even though improvements over previous baselines have been modest in general, the recent progress indicates that deep learning methods will provide valuable means for speeding up or aiding human investigation. Though progress has been made linking a specific neural network's prediction to input features, understanding how users should interpret these models to make testable hypotheses about the system under study remains an open challenge. Furthermore, the limited amount of labelled data for training presents problems in some domains, as do legal and privacy constraints on work with sensitive health records. Nonetheless, we foresee deep learning enabling changes at both bench and bedside with the potential to transform several areas of biology and medicine.

Generating Focused Molecule Libraries for Drug Discovery with Recurrent Neural Networks

In de novo drug design, computational strategies are used to generate novel molecules with good affinity to the desired biological target. In this work, we show that recurrent neural networks can be trained as generative models for molecular structures, similar to statistical language models in natural language processing. We demonstrate that the properties of the generated molecules correlate very well with the properties of the molecules used to train the model. In order to enrich libraries with molecules active toward a given biological target, we propose to fine-tune the model with small sets of molecules, which are known to be active against that target. Against Staphylococcus aureus, the model reproduced 14% of 6051 hold-out test molecules that medicinal chemists designed, whereas against Plasmodium falciparum (Malaria), it reproduced 28% of 1240 test molecules. When coupled with a scoring function, our model can perform the complete de novo drug design cycle to generate large sets of novel molecules for drug discovery.

De Novo Design of Ligands Using Computational Methods

De novo design technique is complementary to high-throughput virtual screening and is believed to contribute in pharmaceutical development of novel drugs with desired properties at a very low cost and time-efficient manner. In this chapter, we outline the basic de novo design concepts based on computational methods with an example.

Automating drug discovery

Small-molecule drug discovery can be viewed as a challenging multidimensional problem in which various characteristics of compounds - including efficacy, pharmacokinetics and safety - need to be optimized in parallel to provide drug candidates. Recent advances in areas such as microfluidics-assisted chemical synthesis and biological testing, as well as artificial intelligence systems that improve a design hypothesis through feedback analysis, are now providing a basis for the introduction of greater automation into aspects of this process. This could potentially accelerate time frames for compound discovery and optimization and enable more effective searches of chemical space. However, such approaches also raise considerable conceptual, technical and organizational challenges, as well as scepticism about the current hype around them. This article aims to identify the approaches and technologies that could be implemented robustly by medicinal chemists in the near future and to critically analyse the opportunities and challenges for their more widespread application.

Neural-Symbolic Machine Learning for Retrosynthesis and Reaction Prediction

Reaction prediction and retrosynthesis are the cornerstones of organic chemistry. Rule-based expert systems have been the most widespread approach to computationally solve these two related challenges to date. However, reaction rules often fail because they ignore the molecular context, which leads to reactivity conflicts. Herein, we report that deep neural networks can learn to resolve reactivity conflicts and to prioritize the most suitable transformation rules. We show that by training our model on 3.5 million reactions taken from the collective published knowledge of the entire discipline of chemistry, our model exhibits a top10-accuracy of 95 % in retrosynthesis and 97 % for reaction prediction on a validation set of almost 1 million reactions.

Scoring of de novo Designed Chemical Entities by Macromolecular Target Prediction

Computational de novo molecular design and macromolecular target prediction have become routine in applied cheminformatics. In this study, we have generated populations of drug template-derived designs using ligand-based building block assembly, and predicted their potential targets. The results of our analysis show that the reaction-based de novo design generated new chemical entities with similar properties and pharmacophores as that of the template drugs as well as up to 44 % of the de novo compounds receiving the correct target predictions. Keeping in mind the probabilistic nature of the methods, such a combination of fast and meaningful computational structure generation by reaction-based design and product scoring by target class prediction may be appropriate for prospective application in medicinal chemistry.

A Pareto Algorithm for Efficient De Novo Design of Multi-functional Molecules

We have introduced a Pareto sorting algorithm into Synopsis, a de novo design program that generates synthesizable molecules with desirable properties. We give a detailed description of the algorithm and illustrate its working in 2 different de novo design settings: the design of putative dual and selective FGFR and VEGFR inhibitors, and the successful design of organic structure determining agents (OSDAs) for the synthesis of zeolites. We show that the introduction of Pareto sorting not only enables the simultaneous optimization of multiple properties but also greatly improves the performance of the algorithm to generate molecules with hard-to-meet constraints. This in turn allows us to suggest approaches to address the problem of false positive hits in de novo structure based drug design by introducing structural and physicochemical constraints in the designed molecules, and by forcing essential interactions between these molecules and their target receptor.

De novo design - hop(p)ing against hope

Current trends in computational de novo design provide a fresh approach to 'scaffold-hopping' in drug discovery. The methodological repertoire is no longer limited to receptor-based methods, but specifically ligand-based techniques that consider multiple properties in parallel, including the synthetic feasibility of the computer-generated molecules and their polypharmacology, provide innovative ideas for the discovery of new chemical entities. The concept of fragment-based and virtual reaction-driven design enables rapid compound optimization from scratch with a manageable complexity of the search. Starting from known drugs as a reference, such algorithms suggest drug-like molecules with motivated scaffold variations, and advanced mathematical models of structure-activity landscapes and multi-objective design techniques have created new opportunities for hit and lead finding.

Future De Novo Drug Design

The computer-assisted generation of new chemical entities (NCEs) has matured into solid technology supporting early drug discovery. Both ligand- and receptor-based methods are increasingly used for designing small lead- and druglike molecules with anticipated multi-target activities. Advanced "polypharmacology" prediction tools are essential pillars of these endeavors. In addition, it has been realized that iterative design-synthesis-test cycles facilitate the rapid identification of NCEs with the desired activity profile. Lab-on-a-chip platforms integrating synthesis, analytics and bioactivity determination and controlled by adaptive, chemistry-driven de novo design software will play an important role for future drug discovery.

Advances in multiparameter optimization methods for de novo drug design

INTRODUCTION

A high-quality drug must achieve a balance of physicochemical and absorption, distribution, metabolism and elimination properties, safety and potency against its therapeutic target(s). Multiparameter optimization (MPO) methods guide the simultaneous optimization of multiple factors to quickly target compounds with the highest chance of downstream success. MPO can be combined with 'de novo design' methods to automatically generate and assess a large number of diverse structures and identify strategies to optimize a compound's overall balance of properties.

AREAS COVERED

The article provides a review of MPO methods and recent developments in the methods and opinions in the field. It also provides a description of advances in de novo design that improve the relevance of automatically generated compound structures and integrate MPO. Finally, the article provides discussion of a recent case study of the automatic design of ligands to polypharmacological profiles.

EXPERT OPINION

Recent developments have reduced the generation of chemically infeasible structures and improved the quality of compounds generated by de novo design methods. There are concerns about the ability of simple drug-like properties and ligand efficiency indices to effectively guide the detailed optimization of compounds. De novo design methods cannot identify a perfect compound for synthesis, but it can identify high-quality ideas for detailed consideration by an expert scientist.

Multi-objective molecular de novo design by adaptive fragment prioritization

We present the development and application of a computational molecular de novo design method for obtaining bioactive compounds with desired on- and off-target binding. The approach translates the nature-inspired concept of ant colony optimization to combinatorial building block selection. By relying on publicly available structure-activity data, we developed a predictive quantitative polypharmacology model for 640 human drug targets. By taking reductive amination as an example of a privileged reaction, we obtained novel subtype-selective and multitarget-modulating dopamine D4 antagonists, as well as ligands selective for the sigma-1 receptor with accurately predicted affinities. The nanomolar potencies of the hits obtained, their high ligand efficiencies, and an overall success rate of 90 % demonstrate that this ligand-based computer-aided molecular design method may guide target-focused combinatorial chemistry.

Drugs by numbers: reaction-driven de novo design of potent and selective anticancer leads

A potent and selective inhibitor of the anticancer target Polo-like kinase 1 was found by computer-based molecular design. This type II kinase inhibitor was synthesized as suggested by the design software DOGS and exhibited significant antiproliferative effects against HeLa cells without affecting nontransformed cells. The study provides a proof-of-concept for reaction-based de novo design as a leading tool for drug discovery.

Target Profile Prediction and Practical Evaluation of a Biginelli-Type Dihydropyrimidine Compound Library

We present a self-organizing map (SOM) approach to predicting macromolecular targets for combinatorial compound libraries. The aim was to study the usefulness of the SOM in combination with a topological pharmacophore representation (CATS) for selecting biologically active compounds from a virtual combinatorial compound collection, taking the multi-component Biginelli dihydropyrimidine reaction as an example. We synthesized a candidate compound from this library, for which the SOM model suggested inhibitory activity against cyclin-dependent kinase 2 (CDK2) and other kinases. The prediction was confirmed in an in vitro panel assay comprising 48 human kinases. We conclude that the computational technique may be used for ligand-based in silico pharmacology studies, off-target prediction, and drug re-purposing, thereby complementing receptor-based approaches.