Generating Multibillion Chemical Space of Readily Accessible Screening Compounds

A Repurposed Drug Interferes with Nucleic Acid to Inhibit the Dual Activities of Coronavirus Nsp13

The recent pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) highlighted a critical need to discover more effective antivirals. While therapeutics for SARS-CoV-2 exist, its nonstructural protein 13 (Nsp13) remains a clinically untapped target. Nsp13 is a helicase responsible for unwinding double-stranded RNA during viral replication and is essential for propagation. Like other helicases, Nsp13 has two active sites: a nucleotide binding site that hydrolyzes nucleoside triphosphates (NTPs) and a nucleic acid binding channel that unwinds double-stranded RNA or DNA. Targeting viral helicases with small molecules, as well as the identification of ligand binding pockets, have been ongoing challenges, partly due to the flexible nature of these proteins. Here, we use a virtual screen to identify ligands of Nsp13 from a collection of clinically used drugs. We find that a known ion channel inhibitor, IOWH-032, inhibits the dual ATPase and helicase activities of SARS-CoV-2 Nsp13 at low micromolar concentrations. Kinetic and binding assays, along with computational and mutational analyses, indicate that IOWH-032 interacts with the RNA binding interface, leading to displacement of nucleic acid substrate, but not bound ATP. Evaluation of IOWH-032 with microbial helicases from other superfamilies reveals that it is selective for coronavirus Nsp13. Furthermore, it remains active against mutants representative of observed SARS-CoV-2 variants. Overall, this work provides a new inhibitor for Nsp13 and provides a rationale for a recent observation that IOWH-032 lowers SARS-CoV-2 viral loads in human cells, setting the stage for the discovery of other potent viral helicase modulators.

Carbenoid-involved reactions integrated with scaffold-based screening generates a Nav1.7 inhibitor

The discovery of selective Nav1.7 inhibitors is a promising approach for developing anti-nociceptive drugs. In this study, we present a novel oxindole-based readily accessible library (OREAL), which is characterized by readily accessibility, unique chemical space, ideal drug-like properties, and structural diversity. We used a scaffold-based approach to screen the OREAL and discovered compound C4 as a potent Nav1.7 inhibitor. The bioactivity characterization of C4 reveals that it is a selective Nav1.7 inhibitor and effectively reverses Paclitaxel-induced neuropathic pain (PINP) in rodent models. Preliminary toxicology study shows C4 is negative to hERG. The consistent results of molecular docking and molecular simulations further support the reasonability of the in-silico screening and show the insight of the binding mode of C4. Our discovery of C4 paves the way for pushing the Nav1.7-based anti-nociceptive drugs forward to the clinic.

Structure-based virtual screening of vast chemical space as a starting point for drug discovery

Structure-based virtual screening aims to find molecules forming favorable interactions with a biological macromolecule using computational models of complexes. The recent surge of commercially available chemical space provides the opportunity to search for ligands of therapeutic targets among billions of compounds. This review offers a compact overview of structure-based virtual screens of vast chemical spaces, highlighting successful applications in early drug discovery for therapeutically important targets such as G protein-coupled receptors and viral enzymes. Emphasis is placed on strategies to explore ultra-large chemical libraries and synergies with emerging machine learning techniques. The current opportunities and future challenges of virtual screening are discussed, indicating that this approach will play an important role in the next-generation drug discovery pipeline.

The Pan-Canadian Chemical Library: A Mechanism to Open Academic Chemistry to High-Throughput Virtual Screening

Computationally screening chemical libraries to discover molecules with desired properties is a common technique used in early-stage drug discovery. Recent progress in the field now enables the efficient exploration of billions of molecules within days or hours, but this exploration remains confined within the boundaries of the accessible chemistry space. While the number of commercially available compounds grows rapidly, it remains a limited subset of all druglike small molecules that could be synthesized. Here, we present a workflow where chemical reactions typically developed in academia and unconventional in drug discovery are exploited to dramatically expand the chemistry space accessible to virtual screening. We use this process to generate a first version of the Pan-Canadian Chemical Library, a collection of nearly 150 billion diverse compounds that does not overlap with other ultra-large libraries such as Enamine REAL or SAVI and could be a resource of choice for protein targets where other libraries have failed to deliver bioactive molecules.

DrugGym: A testbed for the economics of autonomous drug discovery

Drug discovery is stochastic. The effectiveness of candidate compounds in satisfying design objectives is unknown ahead of time, and the tools used for prioritization-predictive models and assays-are inaccurate and noisy. In a typical discovery campaign, thousands of compounds may be synthesized and tested before design objectives are achieved, with many others ideated but deprioritized. These challenges are well-documented, but assessing potential remedies has been difficult. We introduce DrugGym, a framework for modeling the stochastic process of drug discovery. Emulating biochemical assays with realistic surrogate models, we simulate the progression from weak hits to sub-micromolar leads with viable ADME. We use this testbed to examine how different ideation, scoring, and decision-making strategies impact statistical measures of utility, such as the probability of program success within predefined budgets and the expected costs to achieve target candidate profile (TCP) goals. We also assess the influence of affinity model inaccuracy, chemical creativity, batch size, and multi-step reasoning. Our findings suggest that reducing affinity model inaccuracy from 2 to 0.5 pIC50 units improves budget-constrained success rates tenfold. DrugGym represents a realistic testbed for machine learning methods applied to the hit-to-lead phase. Source code is available at www.drug-gym.org.

Target-specific novel molecules with their recipe: Incorporating synthesizability in the design process

Application of Artificial intelligence (AI) in drug discovery has led to several success stories in recent times. While traditional methods mostly relied upon screening large chemical libraries for early-stage drug-design, de novo design can help identify novel target-specific molecules by sampling from a much larger chemical space. Although this has increased the possibility of finding diverse and novel molecules from previously unexplored chemical space, this has also posed a great challenge for medicinal chemists to synthesize at least some of the de novo designed novel molecules for experimental validation. To address this challenge, in this work, we propose a novel forward synthesis-based generative AI method, which is used to explore the synthesizable chemical space. The method uses a structure-based drug design framework, where the target protein structure and a target-specific seed fragment from co-crystal structures can be the initial inputs. A random fragment from a purchasable fragment library can also be the input if a target-specific fragment is unavailable. Then a template-based forward synthesis route prediction and molecule generation is performed in parallel using the Monte Carlo Tree Search (MCTS) method where, the subsequent fragments for molecule growth can again be obtained from a purchasable fragment library. The rewards for each iteration of MCTS are computed using a drug-target affinity (DTA) model based on the docking pose of the generated reaction intermediates at the binding site of the target protein of interest. With the help of the proposed method, it is now possible to overcome one of the major obstacles posed to the AI-based drug design approaches through the ability of the method to design novel target-specific synthesizable molecules.

Correlation of protein binding pocket properties with hits' chemistries used in generation of ultra-large virtual libraries

Although the size of virtual libraries of synthesizable compounds is growing rapidly, we are still enumerating only tiny fractions of the drug-like chemical universe. Our capability to mine these newly generated libraries also lags their growth. That is why fragment-based approaches that utilize on-demand virtual combinatorial libraries are gaining popularity in drug discovery. These à la carte libraries utilize synthetic blocks found to be effective binders in parts of target protein pockets and a variety of reliable chemistries to connect them. There is, however, no data on the potential impact of the chemistries used for making on-demand libraries on the hit rates during virtual screening. There are also no rules to guide in the selection of these synthetic methods for production of custom libraries. We have used the SAVI (Synthetically Accessible Virtual Inventory) library, constructed using 53 reliable reaction types (transforms), to evaluate the impact of these chemistries on docking hit rates for 40 well-characterized protein pockets. The data shows that the virtual hit rates differ significantly for different chemistries with cross coupling reactions such as Sonogashira, Suzuki-Miyaura, Hiyama and Liebeskind-Srogl coupling producing the highest hit rates. Virtual hit rates appear to depend not only on the property of the formed chemical bond but also on the diversity of available building blocks and the scope of the reaction. The data identifies reactions that deserve wider use through increasing the number of corresponding building blocks and suggests the reactions that are more effective for pockets with certain physical and hydrogen bond-forming properties.

Asymmetric Carbene Transfer: Enhancing Chemical Diversity for Drug Discovery

The quest to explore chemical space is vital for identifying novel disease targets, impacting both the effectiveness and safety profile of therapeutic agents. The tangible chemical space, currently estimated at a conservative 108 synthesized compounds, pales in comparison to the theoretically conceivable diversity of 1060 molecules. To bridge this vast gap, organic chemists are spearheading innovative methodologies that promise to broaden this limited chemical diversity. A beacon of this progressive wave is Asymmetric Carbene Transfer (ACT), a burgeoning strategy that significantly boosts molecular diversity with efficient bond-formation and precise chiral control. This review focuses on the capabilities of ACT in creating pharmaceutically significant molecules, encompassing an array of natural products and bioactive compounds. Through the lens of ACT, we discern its substantial influence on drug discovery, paving the way for novel therapeutic avenues by expanding the boundaries of molecular diversity. This review will shed light on prospective methodological developments of ACT and articulate their conceivable contributions to the medicinal chemistry arena.

BitterMasS: Predicting Bitterness from Mass Spectra

Bitter compounds are common in nature and among drugs. Previously, machine learning tools were developed to predict bitterness from the chemical structure. However, known structures are estimated to represent only 5-10% of the metabolome, and the rest remain unassigned or "dark". We present BitterMasS, a Random Forest classifier that was trained on 5414 experimental mass spectra of bitter and nonbitter compounds, achieving precision = 0.83 and recall = 0.90 for an internal test set. Next, the model was tested against spectra newly extracted from the literature 106 bitter and nonbitter compounds and for additional spectra measured for 26 compounds. For these external test cases, BitterMasS exhibited 67% precision and 93% recall for the first and 58% accuracy and 99% recall for the second. The spectrum-bitterness prediction strategy was more effective than the spectrum-structure-bitterness prediction strategy and covered more compounds. These encouraging results suggest that BitterMasS can be used to predict bitter compounds in the metabolome without the need for structural assignment of individual molecules. This may enable identification of bitter compounds from metabolomics analyses, for comparing potential bitterness levels obtained by different treatments of samples and for monitoring bitterness changes overtime.

The landscape of small-molecule prodrugs

Prodrugs are derivatives with superior properties compared with the parent active pharmaceutical ingredient (API), which undergo biotransformation after administration to generate the API in situ. Although sharing this general characteristic, prodrugs encompass a wide range of different chemical structures, therapeutic indications and properties. Here we provide the first holistic analysis of the current landscape of approved prodrugs using cheminformatics and data science approaches to reveal trends in prodrug development. We highlight rationales that underlie prodrug design, their indications, mechanisms of API release, the chemistry of promoieties added to APIs to form prodrugs and the market impact of prodrugs. On the basis of this analysis, we discuss strengths and limitations of current prodrug approaches and suggest areas for future development.

Virtual Screening of a Chemically Diverse "Superscaffold" Library Enables Ligand Discovery for a Key GPCR Target

The advent of ultra-large libraries of drug-like compounds has significantly broadened the possibilities in structure-based virtual screening, accelerating the discovery and optimization of high-quality lead chemotypes for diverse clinical targets. Compared to traditional high-throughput screening, which is constrained to libraries of approximately one million compounds, the ultra-large virtual screening approach offers substantial advantages in both cost and time efficiency. By expanding the chemical space with compounds synthesized from easily accessible and reproducible reactions and utilizing a large, diverse set of building blocks, we can enhance both the diversity and quality of the discovered lead chemotypes. In this study, we explore new chemical spaces using reactions of sulfur(VI) fluorides to create a combinatorial library consisting of several hundred million compounds. We screened this virtual library for cannabinoid type II receptor (CB2) antagonists using the high-resolution structure in conjunction with a rationally designed antagonist, AM10257. The top-predicted compounds were then synthesized and tested in vitro for CB2 binding and functional antagonism, achieving an experimentally validated hit rate of 55%. Our findings demonstrate the effectiveness of reliable reactions, such as sulfur fluoride exchange, in diversifying ultra-large chemical spaces and facilitate the discovery of new lead compounds for important biological targets.

BigBind: Learning from Nonstructural Data for Structure-Based Virtual Screening

Deep learning methods that predict protein-ligand binding have recently been used for structure-based virtual screening. Many such models have been trained using protein-ligand complexes with known crystal structures and activities from the PDBBind data set. However, because PDBbind only includes 20K complexes, models typically fail to generalize to new targets, and model performance is on par with models trained with only ligand information. Conversely, the ChEMBL database contains a wealth of chemical activity information but includes no information about binding poses. We introduce BigBind, a data set that maps ChEMBL activity data to proteins from the CrossDocked data set. BigBind comprises 583 K ligand activities and includes 3D structures of the protein binding pockets. Additionally, we augmented the data by adding an equal number of putative inactives for each target. Using this data, we developed Banana (basic neural network for binding affinity), a neural network-based model to classify active from inactive compounds, defined by a 10 μM cutoff. Our model achieved an AUC of 0.72 on BigBind's test set, while a ligand-only model achieved an AUC of 0.59. Furthermore, Banana achieved competitive performance on the LIT-PCBA benchmark (median EF1% 1.81) while running 16,000 times faster than molecular docking with Gnina. We suggest that Banana, as well as other models trained on this data set, will significantly improve the outcomes of prospective virtual screening tasks.

DFRscore: Deep Learning-Based Scoring of Synthetic Complexity with Drug-Focused Retrosynthetic Analysis for High-Throughput Virtual Screening

Recently emerging generative AI models enable us to produce a vast number of compounds for potential applications. While they can provide novel molecular structures, the synthetic feasibility of the generated molecules is often questioned. To address this issue, a few recent studies have attempted to use deep learning models to estimate the synthetic accessibility of many molecules rapidly. However, retrosynthetic analysis tools used to train the models rely on reaction templates automatically extracted from a large reaction database that are not domain-specific and may exhibit low chemical correctness. To overcome this limitation, we introduce DFRscore (Drug-Focused Retrosynthetic score), a deep learning-based approach for a more practical assessment of synthetic accessibility in drug discovery. The DFRscore model is trained exclusively on drug-focused reactions, providing a predicted number of minimally required synthetic steps for each compound. This approach enables practitioners to filter out compounds that do not meet their desired level of synthetic accessibility at an early stage of high-throughput virtual screening for accelerated drug discovery. The proposed strategy can be easily adapted to other domains by adjusting the synthesis planning setup of the reaction templates and starting materials.

PLANET: A Multi-objective Graph Neural Network Model for Protein-Ligand Binding Affinity Prediction

Predicting protein-ligand binding affinity is a central issue in drug design. Various deep learning models have been published in recent years, where many of them rely on 3D protein-ligand complex structures as input and tend to focus on the single task of reproducing binding affinity. In this study, we have developed a graph neural network model called PLANET (Protein-Ligand Affinity prediction NETwork). This model takes the graph-represented 3D structure of the binding pocket on the target protein and the 2D chemical structure of the ligand molecule as input. It was trained through a multi-objective process with three related tasks, including deriving the protein-ligand binding affinity, protein-ligand contact map, and ligand distance matrix. Besides the protein-ligand complexes with known binding affinity data retrieved from the PDBbind database, a large number of non-binder decoys were also added to the training data for deriving the final model of PLANET. When tested on the CASF-2016 benchmark, PLANET exhibited a scoring power comparable to the best result yielded by other deep learning models as well as a reasonable ranking power and docking power. In virtual screening trials conducted on the DUD-E benchmark, PLANET's performance was notably better than several deep learning and machine learning models. As on the LIT-PCBA benchmark, PLANET achieved comparable accuracy as the conventional docking program Glide, but it only spent less than 1% of Glide's computation time to finish the same job because PLANET did not need exhaustive conformational sampling. Considering the decent accuracy and efficiency of PLANET in binding affinity prediction, it may become a useful tool for conducting large-scale virtual screening.

AI is a viable alternative to high throughput screening: a 318-target study

High throughput screening (HTS) is routinely used to identify bioactive small molecules. This requires physical compounds, which limits coverage of accessible chemical space. Computational approaches combined with vast on-demand chemical libraries can access far greater chemical space, provided that the predictive accuracy is sufficient to identify useful molecules. Through the largest and most diverse virtual HTS campaign reported to date, comprising 318 individual projects, we demonstrate that our AtomNet® convolutional neural network successfully finds novel hits across every major therapeutic area and protein class. We address historical limitations of computational screening by demonstrating success for target proteins without known binders, high-quality X-ray crystal structures, or manual cherry-picking of compounds. We show that the molecules selected by the AtomNet® model are novel drug-like scaffolds rather than minor modifications to known bioactive compounds. Our empirical results suggest that computational methods can substantially replace HTS as the first step of small-molecule drug discovery.

Protein Structure-Based Organic Chemistry-Driven Ligand Design from Ultralarge Chemical Spaces

Ultralarge chemical spaces describing several billion compounds are revolutionizing hit identification in early drug discovery. Because of their size, such chemical spaces cannot be fully enumerated and require ad-hoc computational tools to navigate them and pick potentially interesting hits. We here propose a structure-based approach to ultralarge chemical space screening in which commercial chemical reagents are first docked to the target of interest and then directly connected according to organic chemistry and topological rules, to enumerate drug-like compounds under three-dimensional constraints of the target. When applied to bespoke chemical spaces of different sizes and chemical complexity targeting two receptors of pharmaceutical interest (estrogen β receptor, dopamine D3 receptor), the computational method was able to quickly enumerate hits that were either known ligands (or very close analogs) of targeted receptors as well as chemically novel candidates that could be experimentally confirmed by in vitro binding assays. The proposed approach is generic, can be applied to any docking algorithm, and requires few computational resources to prioritize easily synthesizable hits from billion-sized chemical spaces.

Recent Advances in Automated Structure-Based De Novo Drug Design

As the number of determined and predicted protein structures and the size of druglike 'make-on-demand' libraries soar, the time-consuming nature of structure-based computer-aided drug design calls for innovative computational algorithms. De novo drug design introduces in silico heuristics to accelerate searching in the vast chemical space. This review focuses on recent advances in structure-based de novo drug design, ranging from conventional fragment-based methods, evolutionary algorithms, and Metropolis Monte Carlo methods to deep generative models. Due to the historical limitation of de novo drug design generating readily available drug-like molecules, we highlight the synthetic accessibility efforts in each category and the benchmarking strategies taken to validate the proposed framework.

DrugHIVE: Target-specific spatial drug design and optimization with a hierarchical generative model

Rapid advancement in the computational methods of structure-based drug design has led to their widespread adoption as key tools in the early drug development process. Recently, the remarkable growth of available crystal structure data and libraries of commercially available or readily synthesizable molecules have unlocked previously inaccessible regions of chemical space for drug development. Paired with improvements in virtual ligand screening methods, these expanded libraries are having a significant impact on the success of early drug design efforts. However, screening-based methods are limited in their scalability due to computational limits and the sheer scale of drug-like space. An approach within the quickly evolving field of artificial intelligence (AI), deep generative modeling, is extending the reach of molecular design beyond classical methods by learning the fundamental intra- and inter-molecular relationships in drug-target systems from existing data. In this work we introduce DrugHIVE, a deep hierarchical structure-based generative model that enables fine-grained control over molecular generation. Our model outperforms state of the art autoregressive and diffusion-based methods on common benchmarks and in speed of generation. Here, we demonstrate DrugHIVEs capacity to accelerate a wide range of common drug design tasks such as de novo generation, molecular optimization, scaffold hopping, linker design, and high throughput pattern replacement. Our method is highly scalable and can be applied to high confidence AlphaFold predicted receptors, extending our ability to generate high quality drug-like molecules to a majority of the unsolved human proteome.

Novel Fragment Inhibitors of PYCR1 from Docking-Guided X-ray Crystallography

The proline biosynthetic enzyme Δ1-pyrroline-5-carboxylate (P5C) reductase 1 (PYCR1) is one of the most consistently upregulated enzymes across multiple cancer types and central to the metabolic rewiring of cancer cells. Herein, we describe a fragment-based, structure-first approach to the discovery of PYCR1 inhibitors. Thirty-seven fragment-like carboxylic acids in the molecular weight range of 143-289 Da were selected from docking and then screened using X-ray crystallography as the primary assay. Strong electron density was observed for eight compounds, corresponding to a crystallographic hit rate of 22%. The fragments are novel compared to existing proline analog inhibitors in that they block both the P5C substrate pocket and the NAD(P)H binding site. Four hits showed inhibition of PYCR1 in kinetic assays, and one has lower apparent IC50 than the current best proline analog inhibitor. These results show proof-of-concept for our inhibitor discovery approach and provide a basis for fragment-to-lead optimization.

SNAr or Sulfonylation? Chemoselective Amination of Halo(het)arene Sulfonyl Halides for Synthetic Applications and Ultralarge Compound Library Design

The chemoselectivity of halo(het)arene sulfonyl halide aminations is studied thoroughly under parallel synthesis conditions, and the scope and limitations of the method are established. It is shown that SNAr-reactive sulfonyl halides typically undergo sulfonamide synthesis during the first step; the second amination is also possible provided that the SNAr-active center is sufficiently reactive. On the contrary, sulfonyl fluorides bearing an arylating moiety undergo selective transformation at the latter reactive center under proper control. Further sulfur-fluoride exchange (SuFEx) is also possible, which can be especially valuable for some sulfonyl halide classes. The developed two-step parallel double amination protocol provides access to a 6.67-billion compound synthetically tractable REAL-type chemical space (76% expected synthesis success rate).

Integrating multiscale and machine learning approaches towards the SAMPL9 log P challenge

The partition coefficient (log P) is an important physicochemical property that provides information regarding a molecule's pharmacokinetics, toxicity, and bioavailability. Methods to accurately predict the partition coefficient have the potential to accelerate drug design. In an effort to test current methods and explore new computational techniques, the statistical assessment of the modeling of proteins and ligands (SAMPL) has established a blind prediction challenge. The ninth iteration challenge was to predict the toluene-water partition coefficient (log Ptol/w) of sixteen drug molecules. Herein, three approaches are reported broadly under the categories of quantum mechanics (QM), molecular mechanics (MM), and data-driven machine learning (ML). The three blind submissions yield mean unsigned errors (MUE) ranging from 1.53-2.93 log Ptol/w units. The MUEs were reduced to 1.00 log Ptol/w for the QM methods. While MM and ML methods outperformed DFT approaches for challenge molecules with fewer rotational degrees of freedom, they suffered for the larger molecules in this dataset. Overall, DFT functionals paired with a triple-ζ basis set were the simplest and most effective tool to obtain quantitatively accurate partition coefficients.

COATI: Multimodal Contrastive Pretraining for Representing and Traversing Chemical Space

Creating a successful small molecule drug is a challenging multiparameter optimization problem in an effectively infinite space of possible molecules. Generative models have emerged as powerful tools for traversing data manifolds composed of images, sounds, and text and offer an opportunity to dramatically improve the drug discovery and design process. To create generative optimization methods that are more useful than brute-force molecular generation and filtering via virtual screening, we propose that four integrated features are necessary: large, quantitative data sets of molecular structure and activity, an invertible vector representation of realistic accessible molecules, smooth and differentiable regressors that quantify uncertainty, and algorithms to simultaneously optimize properties of interest. Over the course of 12 months, Terray Therapeutics has collected a data set of 2 billion quantitative binding measurements of small molecules to therapeutic targets, which directly motivates multiparameter generative optimization of molecules conditioned on these data. To this end, we present contrastive optimization for accelerated therapeutic inference (COATI), a pretrained, multimodal encoder-decoder model of druglike chemical space. COATI is constructed without any human biasing of features, using contrastive learning from text and 3D representations of molecules to allow for downstream use with structural models. We demonstrate that COATI possesses many of the desired properties of universal molecular embedding: fixed-dimension, invertibility, autoencoding, accurate regression, and low computation cost. Finally, we present a novel metadynamics algorithm for generative optimization using a small subset of our proprietary data collected for a model protein, carbonic anhydrase, designing molecules that satisfy the multiparameter optimization task of potency, solubility, and drug likeness. This work sets the stage for fully integrated generative molecular design and optimization for small molecules.

Curation and cheminformatics analysis of a Ugi-reaction derived library (URDL) of synthetically tractable small molecules for virtual screening application

Virtual screening (VS) is an important approach in drug discovery and relies on the availability of a virtual library of synthetically tractable molecules. Ugi reaction (UR) represents an important multi-component reaction (MCR) that reliably produces a peptidomimetic scaffold. Recent literature shows that a tactically assembled Ugi adduct can be subjected to further chemical modifications to yield a variety of rings and scaffolds, thus, renewing the interest in this old reaction. Given the reliability and efficiency of UR, we collated an UR derived library (URDL) of small molecules (total = 5773) for VS. The synthesis of the majority of URDL molecules may be carried out in 1-2 pots in a time and cost-effective manner. The detailed analysis of the average property and chemical space of URDL was also carried out using the open-source Datawarrior program. The comparison with FDA-approved oral drugs and inhibitors of protein-protein interactions (iPPIs) suggests URDL molecules are 'clean', drug-like, and conform to a structurally distinct space from the other two categories. The average physicochemical properties of compounds in the URDL library lie closer to iPPI molecules than oral drugs thus suggesting that the URDL resource can be applied to discover novel iPPI molecules. The URDL molecules consist of diverse ring systems, many of which have not been exploited yet for drug design. Thus, URDL represents a small virtual library of drug-like molecules with unexplored chemical space designed for VS. The structures of all molecules of URDL, oral drugs, and iPPI compounds are being made freely accessible as supplementary information for broader application.

Targeting ion channels with ultra-large library screening for hit discovery

Ion channels play a crucial role in a variety of physiological and pathological processes, making them attractive targets for drug development in diseases such as diabetes, epilepsy, hypertension, cancer, and chronic pain. Despite the importance of ion channels in drug discovery, the vastness of chemical space and the complexity of ion channels pose significant challenges for identifying drug candidates. The use of in silico methods in drug discovery has dramatically reduced the time and cost of drug development and has the potential to revolutionize the field of medicine. Recent advances in computer hardware and software have enabled the screening of ultra-large compound libraries. Integration of different methods at various scales and dimensions is becoming an inevitable trend in drug development. In this review, we provide an overview of current state-of-the-art computational chemistry methodologies for ultra-large compound library screening and their application to ion channel drug discovery research. We discuss the advantages and limitations of various in silico techniques, including virtual screening, molecular mechanics/dynamics simulations, and machine learning-based approaches. We also highlight several successful applications of computational chemistry methodologies in ion channel drug discovery and provide insights into future directions and challenges in this field.

Design and Diversity Analysis of Chemical Libraries in Drug Discovery

Chemical libraries and compound data sets are among the main inputs to start the drug discovery process at universities, research institutes, and the pharmaceutical industry. The approach used in the design of compound libraries, the chemical information they possess, and the representation of structures, play a fundamental role in the development of studies: chemoinformatics, food informatics, in silico pharmacokinetics, computational toxicology, bioinformatics, and molecular modeling to generate computational hits that will continue the optimization process of drug candidates. The prospects for growth in drug discovery and development processes in chemical, biotechnological, and pharmaceutical companies began a few years ago by integrating computational tools with artificial intelligence methodologies. It is anticipated that it will increase the number of drugs approved by regulatory agencies shortly.

Docking for EP4R antagonists active against inflammatory pain

The lipid prostaglandin E2 (PGE2) mediates inflammatory pain by activating G protein-coupled receptors, including the prostaglandin E2 receptor 4 (EP4R). Nonsteroidal anti-inflammatory drugs (NSAIDs) reduce nociception by inhibiting prostaglandin synthesis, however, the disruption of upstream prostanoid biosynthesis can lead to pleiotropic effects including gastrointestinal bleeding and cardiac complications. In contrast, by acting downstream, EP4R antagonists may act specifically as anti-inflammatory agents and, to date, no selective EP4R antagonists have been approved for human use. In this work, seeking to diversify EP4R antagonist scaffolds, we computationally dock over 400 million compounds against an EP4R crystal structure and experimentally validate 71 highly ranked, de novo synthesized molecules. Further, we show how structure-based optimization of initial docking hits identifies a potent and selective antagonist with 16 nanomolar potency. Finally, we demonstrate favorable pharmacokinetics for the discovered compound as well as anti-allodynic and anti-inflammatory activity in several preclinical pain models in mice.

A practical guide to machine-learning scoring for structure-based virtual screening

Structure-based virtual screening (SBVS) via docking has been used to discover active molecules for a range of therapeutic targets. Chemical and protein data sets that contain integrated bioactivity information have increased both in number and in size. Artificial intelligence and, more concretely, its machine-learning (ML) branch, including deep learning, have effectively exploited these data sets to build scoring functions (SFs) for SBVS against targets with an atomic-resolution 3D model (e.g., generated by X-ray crystallography or predicted by AlphaFold2). Often outperforming their generic and non-ML counterparts, target-specific ML-based SFs represent the state of the art for SBVS. Here, we present a comprehensive and user-friendly protocol to build and rigorously evaluate these new SFs for SBVS. This protocol is organized into four sections: (i) using a public benchmark of a given target to evaluate an existing generic SF; (ii) preparing experimental data for a target from public repositories; (iii) partitioning data into a training set and a test set for subsequent target-specific ML modeling; and (iv) generating and evaluating target-specific ML SFs by using the prepared training-test partitions. All necessary code and input/output data related to three example targets (acetylcholinesterase, HMG-CoA reductase, and peroxisome proliferator-activated receptor-α) are available at https://github.com/vktrannguyen/MLSF-protocol , can be run by using a single computer within 1 week and make use of easily accessible software/programs (e.g., Smina, CNN-Score, RF-Score-VS and DeepCoy) and web resources. Our aim is to provide practical guidance on how to augment training data to enhance SBVS performance, how to identify the most suitable supervised learning algorithm for a data set, and how to build an SF with the highest likelihood of discovering target-active molecules within a given compound library.

Expanding Bioactive Fragment Space with the Generated Database GDB-13s

Identifying innovative fragments for drug design can help medicinal chemistry address new targets and overcome the limitations of the classical molecular series. By deconstructing molecules into ring fragments (RFs, consisting of ring atoms plus ring-adjacent atoms) and acyclic fragments (AFs, consisting of only acyclic atoms), we find that public databases of molecules (i.e., ZINC and PubChem) and natural products (i.e., COCONUT) contain mostly RFs and AFs of up to 13 atoms. We also find that many RFs and AFs are enriched in bioactive vs inactive compounds from ChEMBL. We then analyze the generated database GDB-13s, which enumerates 99 million possible molecules of up to 13 atoms, for RFs and AFs resembling ChEMBL bioactive RFs and AFs. This analysis reveals a large number of novel RFs and AFs that are structurally simple, have favorable synthetic accessibility scores, and represent opportunities for synthetic chemistry to contribute to drug innovation in the context of fragment-based drug discovery.

Recent Developments in Ultralarge and Structure-Based Virtual Screening Approaches

Drug development is a wide scientific field that faces many challenges these days. Among them are extremely high development costs, long development times, and a small number of new drugs that are approved each year. New and innovative technologies are needed to solve these problems that make the drug discovery process of small molecules more time and cost efficient, and that allow previously undruggable receptor classes to be targeted, such as protein-protein interactions. Structure-based virtual screenings (SBVSs) have become a leading contender in this context. In this review, we give an introduction to the foundations of SBVSs and survey their progress in the past few years with a focus on ultralarge virtual screenings (ULVSs). We outline key principles of SBVSs, recent success stories, new screening techniques, available deep learning-based docking methods, and promising future research directions. ULVSs have an enormous potential for the development of new small-molecule drugs and are already starting to transform early-stage drug discovery.

Sequence-based drug design as a concept in computational drug design

Drug development based on target proteins has been a successful approach in recent decades. However, the conventional structure-based drug design (SBDD) pipeline is a complex, human-engineered process with multiple independently optimized steps. Here, we propose a sequence-to-drug concept for computational drug design based on protein sequence information by end-to-end differentiable learning. We validate this concept in three stages. First, we design TransformerCPI2.0 as a core tool for the concept, which demonstrates generalization ability across proteins and compounds. Second, we interpret the binding knowledge that TransformerCPI2.0 learned. Finally, we use TransformerCPI2.0 to discover new hits for challenging drug targets, and identify new target for an existing drug based on an inverse application of the concept. Overall, this proof-of-concept study shows that the sequence-to-drug concept adds a perspective on drug design. It can serve as an alternative method to SBDD, particularly for proteins that do not yet have high-quality 3D structures available.

Therapeutic disruption of RAD52-ssDNA complexation via novel drug-like inhibitors

RAD52 protein is a coveted target for anticancer drug discovery. Similar to poly-ADP-ribose polymerase (PARP) inhibitors, pharmacological inhibition of RAD52 is synthetically lethal with defects in genome caretakers BRCA1 and BRCA2 (∼25% of breast and ovarian cancers). Emerging structure activity relationships for RAD52 are complex, making it challenging to transform previously identified disruptors of the RAD52-ssDNA interaction into drug-like leads using traditional medicinal chemistry approaches. Using pharmacophoric informatics on the RAD52 complexation by epigallocatechin (EGC), and the Enamine in silico REAL database, we identified six distinct chemical scaffolds that occupy the same physical space on RAD52 as EGC. All six were RAD52 inhibitors (IC50 ∼23-1200 μM) with two of the compounds (Z56 and Z99) selectively killing BRCA-mutant cells and inhibiting cellular activities of RAD52 at micromolar inhibitor concentrations. While Z56 had no effect on the ssDNA-binding protein RPA and was toxic to BRCA-mutant cells only, Z99 inhibited both proteins and displayed toxicity towards BRCA-complemented cells. Optimization of the Z99 scaffold resulted in a set of more powerful and selective inhibitors (IC50 ∼1.3-8 μM), which were only toxic to BRCA-mutant cells. RAD52 complexation by Z56, Z99 and its more specific derivatives provide a roadmap for next generation of cancer therapeutics.

AI-Assisted chemical probe discovery for the understudied Calcium-Calmodulin Dependent Kinase, PNCK

PNCK, or CAMK1b, is an understudied kinase of the calcium-calmodulin dependent kinase family which recently has been identified as a marker of cancer progression and survival in several large-scale multi-omics studies. The biology of PNCK and its relation to oncogenesis has also begun to be elucidated, with data suggesting various roles in DNA damage response, cell cycle control, apoptosis and HIF-1-alpha related pathways. To further explore PNCK as a clinical target, potent small-molecule molecular probes must be developed. Currently, there are no targeted small molecule inhibitors in pre-clinical or clinical studies for the CAMK family. Additionally, there exists no experimentally derived crystal structure for PNCK. We herein report a three-pronged chemical probe discovery campaign which utilized homology modeling, machine learning, virtual screening and molecular dynamics to identify small molecules with low-micromolar potency against PNCK activity from commercially available compound libraries. We report the discovery of a hit-series for the first targeted effort towards discovering PNCK inhibitors that will serve as the starting point for future medicinal chemistry efforts for hit-to-lead optimization of potent chemical probes.

Computational approaches streamlining drug discovery

Computer-aided drug discovery has been around for decades, although the past few years have seen a tectonic shift towards embracing computational technologies in both academia and pharma. This shift is largely defined by the flood of data on ligand properties and binding to therapeutic targets and their 3D structures, abundant computing capacities and the advent of on-demand virtual libraries of drug-like small molecules in their billions. Taking full advantage of these resources requires fast computational methods for effective ligand screening. This includes structure-based virtual screening of gigascale chemical spaces, further facilitated by fast iterative screening approaches. Highly synergistic are developments in deep learning predictions of ligand properties and target activities in lieu of receptor structure. Here we review recent advances in ligand discovery technologies, their potential for reshaping the whole process of drug discovery and development, as well as the challenges they encounter. We also discuss how the rapid identification of highly diverse, potent, target-selective and drug-like ligands to protein targets can democratize the drug discovery process, presenting new opportunities for the cost-effective development of safer and more effective small-molecule treatments.

SARS-CoV2 billion-compound docking

This dataset contains ligand conformations and docking scores for 1.4 billion molecules docked against 6 structural targets from SARS-CoV2, representing 5 unique proteins: MPro, NSP15, PLPro, RDRP, and the Spike protein. Docking was carried out using the AutoDock-GPU platform on the Summit supercomputer and Google Cloud. The docking procedure employed the Solis Wets search method to generate 20 independent ligand binding poses per compound. Each compound geometry was scored using the AutoDock free energy estimate, and rescored using RFScore v3 and DUD-E machine-learned rescoring models. Input protein structures are included, suitable for use by AutoDock-GPU and other docking programs. As the result of an exceptionally large docking campaign, this dataset represents a valuable resource for discovering trends across small molecule and protein binding sites, training AI models, and comparing to inhibitor compounds targeting SARS-CoV-2. The work also gives an example of how to organize and process data from ultra-large docking screens.

Small molecule metabolites: discovery of biomarkers and therapeutic targets

Metabolic abnormalities lead to the dysfunction of metabolic pathways and metabolite accumulation or deficiency which is well-recognized hallmarks of diseases. Metabolite signatures that have close proximity to subject's phenotypic informative dimension, are useful for predicting diagnosis and prognosis of diseases as well as monitoring treatments. The lack of early biomarkers could lead to poor diagnosis and serious outcomes. Therefore, noninvasive diagnosis and monitoring methods with high specificity and selectivity are desperately needed. Small molecule metabolites-based metabolomics has become a specialized tool for metabolic biomarker and pathway analysis, for revealing possible mechanisms of human various diseases and deciphering therapeutic potentials. It could help identify functional biomarkers related to phenotypic variation and delineate biochemical pathways changes as early indicators of pathological dysfunction and damage prior to disease development. Recently, scientists have established a large number of metabolic profiles to reveal the underlying mechanisms and metabolic networks for therapeutic target exploration in biomedicine. This review summarized the metabolic analysis on the potential value of small-molecule candidate metabolites as biomarkers with clinical events, which may lead to better diagnosis, prognosis, drug screening and treatment. We also discuss challenges that need to be addressed to fuel the next wave of breakthroughs.

Pharmacological Chaperones and Protein Conformational Diseases: Approaches of Computational Structural Biology

Whenever a protein fails to fold into its native structure, a profound detrimental effect is likely to occur, and a disease is often developed. Protein conformational disorders arise when proteins adopt abnormal conformations due to a pathological gene variant that turns into gain/loss of function or improper localization/degradation. Pharmacological chaperones are small molecules restoring the correct folding of a protein suitable for treating conformational diseases. Small molecules like these bind poorly folded proteins similarly to physiological chaperones, bridging non-covalent interactions (hydrogen bonds, electrostatic interactions, and van der Waals contacts) loosened or lost due to mutations. Pharmacological chaperone development involves, among other things, structural biology investigation of the target protein and its misfolding and refolding. Such research can take advantage of computational methods at many stages. Here, we present an up-to-date review of the computational structural biology tools and approaches regarding protein stability evaluation, binding pocket discovery and druggability, drug repurposing, and virtual ligand screening. The tools are presented as organized in an ideal workflow oriented at pharmacological chaperones' rational design, also with the treatment of rare diseases in mind.

Targeting in silico GPCR conformations with ultra-large library screening for hit discovery

The use of deep machine learning (ML) in protein structure prediction has made it possible to easily access a large number of annotated conformations that can potentially compensate for missing experimental structures in structure-based drug discovery (SBDD). However, it is still unclear whether the accuracy of these predicted conformations is sufficient for screening chemical compounds that will effectively interact with a protein target for pharmacological purposes. In this opinion article, we examine the potential benefits and limitations of using state-annotated conformations for ultra-large library screening (ULLS) in light of the growing size of ultra-large libraries (ULLs). We believe that targeting different conformational states of common drug targets like G-protein-coupled receptors (GPCRs), which can regulate human physiology by switching between different conformations, can offer multiple advantages.

Advances in the discovery of new chemotypes through ultra-large library docking

INTRODUCTION

The size and complexity of virtual screening libraries in drug discovery have skyrocketed in recent years, reaching up to multiple billions of accessible compounds. However, virtual screening of such ultra-large libraries poses several challenges associated with preparing the libraries, sampling, and pre-selection of suitable compounds. The utilization of artificial intelligence (AI)-assisted screening approaches, such as deep learning, poses a promising countermeasure to deal with this rapidly expanding chemical space. For example, various AI-driven methods were recently successfully used to identify novel small molecule inhibitors of the SARS-CoV-2 main protease (M^pro).

AREAS COVERED

This review focuses on presenting various kinds of virtual screening methods suitable for dealing with ultra-large libraries. Challenges associated with these computational methodologies are discussed, and recent advances are highlighted in the example of the discovery of novel M^pro inhibitors targeting the SARS-CoV-2 virus.

EXPERT OPINION

With the rapid expansion of the virtual chemical space, the methodologies for docking and screening such quantities of molecules need to keep pace. Employment of AI-driven screening compounds has already been shown to be effective in a range from a few thousand to multiple billion compounds, furthered by de novo generation of drug-like molecules without human interference.

Development of Pharmacophore Models for the Important Off-Target 5-HT2B Receptor

Toxicity is a major cause of attrition in the development of pharmaceuticals, and the off-target effects are a frequent contributor. The 5-HT_2B receptor agonism is known to be responsible for a variety of safety concerns including valvular heart disease which was the cause for the withdrawal of several compounds from the market. An early detection of potential binding to this receptor is thus desirable. Herein, we present the identification of key amino acid residues in the active site of 5-HT_2B by molecular dynamics simulations, the development of pharmacophore models and their performance on in-house data, and a structurally highly diverse subset of Enamine REAL labeled for 5-HT_2B activity by a machine learning model. These models may be used as filters employed on screening compound sets for the early filtration of compounds with potential 5-HT_2B off-target liabilities.

Active Learning Guided Drug Design Lead Optimization Based on Relative Binding Free Energy Modeling

In silico identification of potent protein inhibitors commonly requires prediction of a ligand binding free energy (BFE). Thermodynamics integration (TI) based on molecular dynamics (MD) simulations is a BFE calculation method capable of acquiring accurate BFE, but it is computationally expensive and time-consuming. In this work, we have developed an efficient automated workflow for identifying compounds with the lowest BFE among thousands of congeneric ligands, which requires only hundreds of TI calculations. Automated machine learning (AutoML) orchestrated by active learning (AL) in an AL-AutoML workflow allows unbiased and efficient search for a small set of best-performing molecules. We have applied this workflow to select inhibitors of the SARS-CoV-2 papain-like protease and were able to find 133 compounds with improved binding affinity, including 16 compounds with better than 100-fold binding affinity improvement. We obtained a hit rate that outperforms that expected of traditional expert medicinal chemist-guided campaigns. Thus, we demonstrate that the combination of AL and AutoML with free energy simulations provides at least 20× speedup relative to the naïve brute force approaches.

Rationally designed inhibitors of the Musashi protein-RNA interaction by hotspot mimicry

RNA-binding proteins (RBPs) are key post-transcriptional regulators of gene expression, and thus underlie many important biological processes. Here, we developed a strategy that entails extracting a "hotspot pharmacophore" from the structure of a protein-RNA complex, to create a template for designing small-molecule inhibitors and for exploring the selectivity of the resulting inhibitors. We demonstrate this approach by designing inhibitors of Musashi proteins MSI1 and MSI2, key regulators of mRNA stability and translation that are upregulated in many cancers. We report this novel series of MSI1/MSI2 inhibitors is specific and active in biochemical, biophysical, and cellular assays. This study extends the paradigm of "hotspots" from protein-protein complexes to protein-RNA complexes, supports the "druggability" of RNA-binding protein surfaces, and represents one of the first rationally-designed inhibitors of non-enzymatic RNA-binding proteins. Owing to its simplicity and generality, we anticipate that this approach may also be used to develop inhibitors of many other RNA-binding proteins; we also consider the prospects of identifying potential off-target interactions by searching for other RBPs that recognize their cognate RNAs using similar interaction geometries. Beyond inhibitors, we also expect that compounds designed using this approach can serve as warheads for new PROTACs that selectively degrade RNA-binding proteins.

Rationally designed inhibitors of the Musashi protein-RNA interaction by hotspot mimicry

RNA-binding proteins (RBPs) are key post-transcriptional regulators of gene expression, and thus underlie many important biological processes. Here, we developed a strategy that entails extracting a "hotspot pharmacophore" from the structure of a protein-RNA complex, to create a template for designing small-molecule inhibitors and for exploring the selectivity of the resulting inhibitors. We demonstrate this approach by designing inhibitors of Musashi proteins MSI1 and MSI2, key regulators of mRNA stability and translation that are upregulated in many cancers. We report this novel series of MSI1/MSI2 inhibitors is specific and active in biochemical, biophysical, and cellular assays. This study extends the paradigm of "hotspots" from protein-protein complexes to protein-RNA complexes, supports the "druggability" of RNA-binding protein surfaces, and represents one of the first rationally-designed inhibitors of non-enzymatic RNA-binding proteins. Owing to its simplicity and generality, we anticipate that this approach may also be used to develop inhibitors of many other RNA-binding proteins; we also consider the prospects of identifying potential off-target interactions by searching for other RBPs that recognize their cognate RNAs using similar interaction geometries. Beyond inhibitors, we also expect that compounds designed using this approach can serve as warheads for new PROTACs that selectively degrade RNA-binding proteins.

Computer-Aided Drug Design: An Update

Computer-aided drug design (CADD) approaches are playing an increasingly important role in understanding the fundamentals of ligand-receptor interactions and helping medicinal chemists design therapeutics. About 5 years ago, we presented a chapter devoted to an overview of CADD methods and covered typical CADD protocols including structure-based drug design (SBDD) and ligand-based drug design (LBDD) approaches that were frequently used in the antibiotic drug design process. Advances in computational hardware and algorithms and emerging CADD methods are enhancing the accuracy and ability of CADD in drug design and development. In this chapter, an update to our previous chapter is provided with a focus on new CADD approaches from our laboratory and other peers that can be employed to facilitate the development of antibiotic therapeutics.

Overlap of On-demand Ultra-large Combinatorial Spaces with On-the-shelf Drug-like Libraries

On-demand combinatorial spaces are shifting paradigms in early drug discovery, by considerably increasing the searchable chemical space to several billions of compounds while securing their synthetic accessibility. We here systematically compared the on-the-shelf available drug-like chemical space (9 million compounds) to three on-demand ultra-large (ODUL) combinatorial fragment spaces (REAL, CHEMriya, GalaXi) covering 32 billion of readily accessible molecules. Surprisingly, only one space (REAL) intersects almost entirely the currently available drug-like space, suggesting that it is the only ODUL widely suitable for in-stock hit expansion. Of course, expanding a preliminary ODUL hit in the same chemical space is the best possible strategy to rapidly generate structure-activity relationships. All three spaces remain well suited to early hit finding initiatives since they all provide numerous unique scaffolds that are not described by on-the shelf collections.

A vending machine for drug-like molecules - automated synthesis of virtual screening hits

As a result of high false positive rates in virtual screening campaigns, prospective hits must be synthesised for validation. When done manually, this is a time consuming and laborious process. Large "on-demand" virtual libraries (>7 × 10¹² members), suitable for preparation using capsule-based automated synthesis and commercial building blocks, were evaluated to determine their structural novelty. One sub-library, constructed from iSnAP capsules, aldehydes and amines, contains unique scaffolds with drug-like physicochemical properties. Virtual screening hits from this iSnAP library were prepared in an automated fashion for evaluation against Aedes aegypti and Phytophthora infestans. In comparison to manual workflows, this approach provided a 10-fold improvement in user efficiency. A streamlined method of relative stereochemical assignment was also devised to augment the rapid synthesis. User efficiency was further improved to 100-fold by downscaling and parallelising capsule-based chemistry on 96-well plates equipped with filter bases. This work demonstrates that automated synthesis consoles can enable the rapid and reliable preparation of attractive virtual screening hits from large virtual libraries.

The Time and Place for Nature in Drug Discovery

The case for a renewed focus on Nature in drug discovery is reviewed; not in terms of natural product screening, but how and why biomimetic molecules, especially those produced by natural processes, should deliver in the age of artificial intelligence and screening of vast collections both in vitro and in silico. The declining natural product-likeness of licensed drugs and the consequent physicochemical implications of this trend in the context of current practices are noted. To arrest these trends, the logic of seeking new bioactive agents with enhanced natural mimicry is considered; notably that molecules constructed by proteins (enzymes) are more likely to interact with other proteins (e.g., targets and transporters), a notion validated by natural products. Nature's finite number of building blocks and their interactions necessarily reduce potential numbers of structures, yet these enable expansion of chemical space with their inherent diversity of physical characteristics, pertinent to property-based design. The feasible variations on natural motifs are considered and expanded to encompass pseudo-natural products, leading to the further logical step of harnessing bioprocessing routes to access them. Together, these offer opportunities for enhancing natural mimicry, thereby bringing innovation to drug synthesis exploiting the characteristics of natural recognition processes. The potential for computational guidance to help identifying binding commonalities in the route map is a logical opportunity to enable the design of tailored molecules, with a focus on "organic/biological" rather than purely "synthetic" structures. The design and synthesis of prototype structures should pay dividends in the disposition and efficacy of the molecules, while inherently enabling greener and more sustainable manufacturing techniques.

Generative and reinforcement learning approaches for the automated de novo design of bioactive compounds

Deep generative neural networks have been used increasingly in computational chemistry for de novo design of molecules with desired properties. Many deep learning approaches employ reinforcement learning for optimizing the target properties of the generated molecules. However, the success of this approach is often hampered by the problem of sparse rewards as the majority of the generated molecules are expectedly predicted as inactives. We propose several technical innovations to address this problem and improve the balance between exploration and exploitation modes in reinforcement learning. In a proof-of-concept study, we demonstrate the application of the deep generative recurrent neural network architecture enhanced by several proposed technical tricks to design inhibitors of the epidermal growth factor (EGFR) and further experimentally validate their potency. The proposed technical solutions are expected to substantially improve the success rate of finding novel bioactive compounds for specific biological targets using generative and reinforcement learning approaches.