Fragment Database FDB-17

Efficient Exploration of Adsorption Space for Separations in Metal-Organic Frameworks Combining the Use of Molecular Simulations, Machine Learning, and Ideal Adsorbed Solution Theory

Adsorption-based separations using metal-organic frameworks (MOFs) are promising candidates for replacing common energy-intensive separation processes. The so-called adsorption space formed by the combination of billions of possible molecules and thousands of reported MOFs is vast. It is very challenging to comprehensively evaluate the performance of MOFs for chemical separation through experiments. Molecular simulations and machine learning (ML) have been widely applied to make predictions for adsorption-based separations. Previous ML approaches to these issues were typically limited to smaller molecules and often had poor accuracy in the dilute limit. To enable exploration of a wider adsorption space, we carefully selected a diverse set of 45 molecules and 335 MOFs and generated single-component isotherms of 15,075 MOF-molecule pairs by grand canonical Monte Carlo. Using this database, we successfully developed accurate (r2 > 0.9) machine learning models predicting adsorption isotherms of diverse molecules in large libraries of MOFs. With this approach, we can efficiently make predictions of large collections of MOFs for arbitrary mixture separations. By combining molecular simulation data and ML predictions with Ideal Adsorbed Solution Theory, we tested the ability of these approaches to make predictions of adsorption selectivity and loading for challenging near-azeotropic mixtures.

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.

Convolutional neural network-based quantitative structure-activity relationship and fingerprint analysis against inhibitors of anthrax lethal factor

Aim: To develop a one-dimensional convolutional neural network-based quantitative structure-activity relationship (1D-CNN-QSAR) model to identify novel anthrax inhibitors and analyze chemical space. Methods: We developed a 1D-CNN-QSAR model to identify novel anthrax inhibitors. Results: The statistical results of the 1D-CNN-QSAR model showed a mean square error of 0.045 and a predicted correlation coefficient of 0.79 for the test set. Further, chemical space analysis showed more than 80% fragment pair similarity, with activity cliffs associated with carboxylic acid, 2-phenylfurans, N-phenyldihydropyrazole, N-phenylpyrrole, furan, 4-methylene-1H-pyrazol-5-one, phenylimidazole, phenylpyrrole and phenylpyrazolidine. Conclusion: These fragments may serve as the basis for developing potent novel drug candidates for anthrax. Finally, we concluded that our proposed 1D-CNN-QSAR model and fingerprint analysis might be used to discover potential anthrax drug candidates.

GPCRLigNet: rapid screening for GPCR active ligands using machine learning

Molecules with bioactivity towards G protein-coupled receptors represent a subset of the vast space of small drug-like molecules. Here, we compare machine learning models, including dilated graph convolutional networks, that conduct binary classification to quickly identify molecules with activity towards G protein-coupled receptors. The models are trained and validated using a large set of over 600,000 active, inactive, and decoy compounds. The best performing machine learning model, dubbed GPCRLigNet, was a surprisingly simple feedforward dense neural network mapping from Morgan fingerprints to activity. Incorporation of GPCRLigNet into a high-throughput virtual screening workflow is demonstrated with molecular docking towards a particular G protein-coupled receptor, the pituitary adenylate cyclase-activating polypeptide receptor type 1. Through rigorous comparison of docking scores for molecules selected with and without using GPCRLigNet, we demonstrate an enrichment of potentially potent molecules using GPCRLigNet. This work provides a proof of principle that GPCRLigNet can effectively hone the chemical search space towards ligands with G protein-coupled receptor activity.

Fragment-based drug discovery-the importance of high-quality molecule libraries

Fragment-based drug discovery (FBDD) is now established as a complementary approach to high-throughput screening (HTS). Contrary to HTS, where large libraries of drug-like molecules are screened, FBDD screens involve smaller and less complex molecules which, despite a low affinity to protein targets, display more 'atom-efficient' binding interactions than larger molecules. Fragment hits can, therefore, serve as a more efficient start point for subsequent optimisation, particularly for hard-to-drug targets. Since the number of possible molecules increases exponentially with molecular size, small fragment libraries allow for a proportionately greater coverage of their respective 'chemical space' compared with larger HTS libraries comprising larger molecules. However, good library design is essential to ensure optimal chemical and pharmacophore diversity, molecular complexity, and physicochemical characteristics. In this review, we describe our views on fragment library design, and on what constitutes a good fragment from a medicinal and computational chemistry perspective. We highlight emerging chemical and computational technologies in FBDD and discuss strategies for optimising fragment hits. The impact of novel FBDD approaches is already being felt, with the recent approval of the covalent KRAS^G12C inhibitor sotorasib highlighting the utility of FBDD against targets that were long considered undruggable.

Artificial Intelligence in De novo Drug Design: Are We Still There?

BACKGROUND

The artificial intelligence (AI)-assisted design of drug candidates with novel structures and desired properties has received significant attention in the recent past, so related areas of forward prediction that aim to discover chemical matters worth synthesizing and further experimental investigation.

OBJECTIVES

The purpose behind developing AI-driven models is to explore the broader chemical space and suggest new drug candidate scaffolds with promising therapeutic value. Moreover, it is anticipated that such AI-based models may not only significantly reduce the cost and time but also decrease the attrition rate of drug candidates that fail to reach the desirable endpoints at the final stages of drug development. In an attempt to develop AI-based models for de novo drug design, numerous methods have been proposed by various study groups by applying machine learning and deep learning algorithms to chemical datasets. However, there are many challenges in obtaining accurate predictions, and real breakthroughs in de novo drug design are still scarce.

METHODS

In this review, we explore the recent trends in developing AI-based models for de novo drug design to assess the current status, challenges, and opportunities in the field.

CONCLUSION

The consistently improved AI algorithms and the abundance of curated training chemical data indicate that AI-based de novo drug design should perform better than the current models. Improvements in the performance are warranted to obtain better outcomes in the form of potential drug candidates, which can perform well in in vivo conditions, especially in the case of more complex diseases.

Interpretable Machine Learning Models for Molecular Design of Tyrosine Kinase Inhibitors Using Variational Autoencoders and Perturbation-Based Approach of Chemical Space Exploration

In the current study, we introduce an integrative machine learning strategy for the autonomous molecular design of protein kinase inhibitors using variational autoencoders and a novel cluster-based perturbation approach for exploration of the chemical latent space. The proposed strategy combines autoencoder-based embedding of small molecules with a cluster-based perturbation approach for efficient navigation of the latent space and a feature-based kinase inhibition likelihood classifier that guides optimization of the molecular properties and targeted molecular design. In the proposed generative approach, molecules sharing similar structures tend to cluster in the latent space, and interpolating between two molecules in the latent space enables smooth changes in the molecular structures and properties. The results demonstrated that the proposed strategy can efficiently explore the latent space of small molecules and kinase inhibitors along interpretable directions to guide the generation of novel family-specific kinase molecules that display a significant scaffold diversity and optimal biochemical properties. Through assessment of the latent-based and chemical feature-based binary and multiclass classifiers, we developed a robust probabilistic evaluator of kinase inhibition likelihood that is specifically tailored to guide the molecular design of novel SRC kinase molecules. The generated molecules originating from LCK and ABL1 kinase inhibitors yielded ~40% of novel and valid SRC kinase compounds with high kinase inhibition likelihood probability values (p > 0.75) and high similarity (Tanimoto coefficient > 0.6) to the known SRC inhibitors. By combining the molecular perturbation design with the kinase inhibition likelihood analysis and similarity assessments, we showed that the proposed molecular design strategy can produce novel valid molecules and transform known inhibitors of different kinase families into potential chemical probes of the SRC kinase with excellent physicochemical profiles and high similarity to the known SRC kinase drugs. The results of our study suggest that task-specific manipulation of a biased latent space may be an important direction for more effective task-oriented and target-specific autonomous chemical design models.

Fragment-Based Drug Discovery against Mycobacteria: The Success and Challenges

The emergence of drug-resistant mycobacteria, including Mycobacterium tuberculosis (Mtb) and non-tuberculous mycobacteria (NTM), poses an increasing global threat that urgently demands the development of new potent anti-mycobacterial drugs. One of the approaches toward the identification of new drugs is fragment-based drug discovery (FBDD), which is the most ingenious among other drug discovery models, such as structure-based drug design (SBDD) and high-throughput screening. Specialized techniques, such as X-ray crystallography, nuclear magnetic resonance spectroscopy, and many others, are part of the drug discovery approach to combat the Mtb and NTM global menaces. Moreover, the primary drawbacks of traditional methods, such as the limited measurement of biomolecular toxicity and uncertain bioavailability evaluation, are successfully overcome by the FBDD approach. The current review focuses on the recognition of fragment-based drug discovery as a popular approach using virtual, computational, and biophysical methods to identify potent fragment molecules. FBDD focuses on designing optimal inhibitors against potential therapeutic targets of NTM and Mtb (PurC, ArgB, MmpL3, and TrmD). Additionally, we have elaborated on the challenges associated with the FBDD approach in the identification and development of novel compounds. Insights into the applications and overcoming the challenges of FBDD approaches will aid in the identification of potential therapeutic compounds to treat drug-sensitive and drug-resistant NTMs and Mtb infections.

Progress and Impact of Latin American Natural Product Databases

Natural products (NPs) are a rich source of structurally novel molecules, and the chemical space they encompass is far from being fully explored. Over history, NPs have represented a significant source of bioactive molecules and have served as a source of inspiration for developing many drugs on the market. On the other hand, computer-aided drug design (CADD) has contributed to drug discovery research, mitigating costs and time. In this sense, compound databases represent a fundamental element of CADD. This work reviews the progress toward developing compound databases of natural origin, and it surveys computational methods, emphasizing chemoinformatic approaches to profile natural product databases. Furthermore, it reviews the present state of the art in developing Latin American NP databases and their practical applications to the drug discovery area.

Graph-based Automated Macro-Molecule Assembly

We present a general molecular framework assembly algorithm that takes a largely arbitrary molecular fragment database and a user-supplied target template graph as input. Automatic assembly of molecular fragments from the database, following a prescribed, user-supplied set of connection rules, then turns the template graph into an actual, chemically reasonable molecular framework. Assembly capabilities of our algorithm are tested by producing several abstract, closed-loop shapes. To indicate a few of many possible application areas we demonstrate a host-guest complex and a road toward catalysis. Postassembly substituent exchange can be used to produce electric fields of desired values at desired points inside the framework or at its surface as a stepping stone toward rationally designed, artificial heterogeneous catalysts.

Exploration of Ultralarge Compound Collections for Drug Discovery

Designing new medicines more cheaply and quickly is tightly linked to the quest of exploring chemical space more widely and efficiently. Chemical space is monumentally large, but recent advances in computer software and hardware have enabled researchers to navigate virtual chemical spaces containing billions of chemical structures. This review specifically concerns collections of many millions or even billions of enumerated chemical structures as well as even larger chemical spaces that are not fully enumerated. We present examples of chemical libraries and spaces and the means used to construct them, and we discuss new technologies for searching huge libraries and for searching combinatorially in chemical space. We also cover space navigation techniques and consider new approaches to de novo drug design and the impact of the "autonomous laboratory" on synthesis of designed compounds. Finally, we summarize some other challenges and opportunities for the future.

Implementation of an AI-assisted fragment-generator in an open-source platform

We recently reported a deep learning model to facilitate fragment library design, which is critical for efficient hit identification. However, our model was implemented in Python. We have now created an implementation in the KNIME graphical pipelining environment which we hope will allow experimentation by users with limited programming knowledge.

We report a deep learning model to facilitate fragment library design, which is critical for efficient hit identification, and an implementation in the KNIME graphical workflow environment which should facilitate a more codeless use.

Discovery of Novel BRD4 Ligand Scaffolds by Automated Navigation of the Fragment Chemical Space

Fragment-based drug discovery (FBDD) is a very effective hit identification method. However, the evolution of fragment hits into suitable leads remains challenging and largely artisanal. Fragment evolution is often scaffold-centric, meaning that its outcome depends crucially on the chemical structure of the starting fragment. Considering that fragment screening libraries cover only a small proportion of the corresponding chemical space, hits should be seen as probes highlighting privileged areas of the chemical space rather than actual starting points. We have developed an automated computational pipeline to mine the chemical space around any specific fragment hit, rapidly finding analogues that share a common interaction motif but are structurally novel and diverse. On a prospective application on the bromodomain-containing protein 4 (BRD4), starting from a known fragment, the platform yields active molecules with nonobvious scaffold changes. The procedure is fast and inexpensive and has the potential to uncover many hidden opportunities in FBDD.

Inhibitors of Human Divalent Metal Transporters DMT1 (SLC11A2) and ZIP8 (SLC39A8) from a GDB-17 Fragment Library

Solute carrier proteins (SLCs) are membrane proteins controlling fluxes across biological membranes and represent an emerging class of drug targets. Here we searched for inhibitors of divalent metal transporters in a library of 1,676 commercially available 3D-shaped fragment-like molecules from the generated database GDB-17, which lists all possible organic molecules up to 17 atoms of C, N, O, S and halogen following simple criteria for chemical stability and synthetic feasibility. While screening against DMT1 (SLC11A2), an iron transporter associated with hemochromatosis and for which only very few inhibitors are known, only yielded two weak inhibitors, our approach led to the discovery of the first inhibitor of ZIP8 (SLC39A8), a zinc transporter associated with manganese homeostasis and osteoarthritis but with no previously reported pharmacology, demonstrating that this target is druggable.

Automated Generation of Novel Fragments Using Screening Data, a Dual SMILES Autoencoder, Transfer Learning and Syntax Correction

Fragment-based hit identification (FBHI) allows proportionately greater coverage of chemical space using fewer molecules than traditional high-throughput screening approaches. However, effectively exploiting this advantage is highly dependent on the library design. Solubility, stability, chemical complexity, chemical/shape diversity, and synthetic tractability for fragment elaboration are all critical aspects, and molecule design remains a time-consuming task for computational and medicinal chemists. Artificial neural networks have attracted considerable attention in automated de novo design applications and could also prove useful for fragment library design. Chemical autoencoders are neural networks consisting of encoder and decoder parts, which respectively compress and decompress molecular representations. The decoder is applied to samples drawn from the space of compressed representations to generate novel molecules that can be scored for properties of interest. Here, we report an autoencoder model using a recurrent neural network architecture, which was trained using 486,565 fragments curated from commercial sources, to simultaneously reconstruct both SMILES and chemical fingerprints. To explore its utility in fragment design, we applied transfer learning to the fingerprint decoder layers to train a classifier using 66 frequent hitter fragments identified from our screening campaigns. Using a particle swarm optimization sampling approach, we compare the performance of this "dual" model to an architecture encoding SMILES only. The dual model produced valid SMILES with improved features, considering a range of properties including aromatic ring counts, heavy atom count, synthetic accessibility, and a new fragment complexity score we term Feature Complexity (FeCo). Additionally, we demonstrate that generative performance is further enhanced by use of a simple syntax-correction procedure during training, in which invalid and undesirable SMILES are spiked into the training set. Finally, we used the syntax-corrected model to generate a library of novel candidate privileged fragments.

Fragment-based drug design facilitates selective kinase inhibitor discovery

Protein kinases (PKs) are important drug targets, but kinases selectivity poses a challenge to protein kinase inhibitors (PKIs) design. Fragment-based drug discovery (FBDD) has achieved great success in the discovery of highly specific PKIs. It makes full use of kinase-fragment interaction in target kinase subpockets to obtain promising selectivity. However, it's difficult to understand the complicated kinase-fragment interaction space, and systemic discussion of these interactions is still lacking. Herein, we introduce the advantages of the FBDD strategy in PKIs design. Key features of the selectivity of kinase-fragment interactions are summarized and analyzed. Some promising PKIs are introduced as case studies to help understand the fragment-to-lead (F2L) optimization process. Novel strategies and technologies for FBDD in PKIs discovery are also outlooked.

Protocol for hit-to-lead optimization of compounds by auto in silico ligand directing evolution (AILDE) approach

Hit-to-lead (H2L) optimization is crucial for drug design, which has become an increasing concern in medicinal chemistry. A virtual screening strategy of auto in silico ligand directing evolution (AILDE) has been developed to yield promising lead compounds rapidly and efficiently. The protocol includes instructions for fragment compound library construction, conformational sampling by molecular dynamics simulation, ligand modification by fragment growing, as well as the binding free energy prediction. For complete details on the use and execution of this protocol, please refer to Wu et al. (2020).

DrugSpaceX: a large screenable and synthetically tractable database extending drug space

One of the most prominent topics in drug discovery is efficient exploration of the vast drug-like chemical space to find synthesizable and novel chemical structures with desired biological properties. To address this challenge, we created the DrugSpaceX (https://drugspacex.simm.ac.cn/) database based on expert-defined transformations of approved drug molecules. The current version of DrugSpaceX contains >100 million transformed chemical products for virtual screening, with outstanding characteristics in terms of structural novelty, diversity and large three-dimensional chemical space coverage. To illustrate its practical application in drug discovery, we used a case study of discoidin domain receptor 1 (DDR1), a kinase target implicated in fibrosis and other diseases, to show DrugSpaceX performing a quick search of initial hit compounds. Additionally, for ligand identification and optimization purposes, DrugSpaceX also provides several subsets for download, including a 10% diversity subset, an extended drug-like subset, a drug-like subset, a lead-like subset, and a fragment-like subset. In addition to chemical properties and transformation instructions, DrugSpaceX can locate the position of transformation, which will enable medicinal chemists to easily integrate strategy planning and protection design.

Artificial Intelligence in Drug Discovery: A Comprehensive Review of Data-driven and Machine Learning Approaches

As expenditure on drug development increases exponentially, the overall drug discovery process requires a sustainable revolution. Since artificial intelligence (AI) is leading the fourth industrial revolution, AI can be considered as a viable solution for unstable drug research and development. Generally, AI is applied to fields with sufficient data such as computer vision and natural language processing, but there are many efforts to revolutionize the existing drug discovery process by applying AI. This review provides a comprehensive, organized summary of the recent research trends in AI-guided drug discovery process including target identification, hit identification, ADMET prediction, lead optimization, and drug repositioning. The main data sources in each field are also summarized in this review. In addition, an in-depth analysis of the remaining challenges and limitations will be provided, and proposals for promising future directions in each of the aforementioned areas.

Pyrazolyl-pyrimidones inhibit the function of human solute carrier protein SLC11A2 (hDMT1) by metal chelation

Solute carrier proteins (SLCs) control fluxes of ions and molecules across biological membranes and represent an emerging class of drug targets. SLC11A2 (hDMT1) mediates intestinal iron uptake and its inhibition might be used to treat iron overload diseases such as hereditary hemochromatosis. Here we report a micromolar (IC50 = 1.1 μM) pyrazolyl-pyrimidone inhibitor of radiolabeled iron uptake in hDMT1 overexpressing HEK293 cells acting by a non-competitive mechanism, which however does not affect the electrophysiological properties of the transporter. Isothermal titration calorimetry, competition with calcein, induced precipitation of radioactive iron and cross inhibition of the unrelated iron transporter SLC39A8 (hZIP8) indicate that inhibition is mediated by metal chelation. Mapping the chemical space of thousands of pyrazolo-pyrimidones and similar 2,2'-diazabiaryls in ChEMBL suggests that their reported activities might partly reflect metal chelation. Such metal chelating groups are not listed in pan-assay interference compounds (PAINS) but should be checked when addressing SLCs.

Transfer Learning for Drug Discovery

The data sets available to train models for in silico drug discovery efforts are often small. Indeed, the sparse availability of labeled data is a major barrier to artificial-intelligence-assisted drug discovery. One solution to this problem is to develop algorithms that can cope with relatively heterogeneous and scarce data. Transfer learning is a type of machine learning that can leverage existing, generalizable knowledge from other related tasks to enable learning of a separate task with a small set of data. Deep transfer learning is the most commonly used type of transfer learning in the field of drug discovery. This Perspective provides an overview of transfer learning and related applications to drug discovery to date. Furthermore, it provides outlooks on the future development of transfer learning for drug discovery.

Visualization of very large high-dimensional data sets as minimum spanning trees

The chemical sciences are producing an unprecedented amount of large, high-dimensional data sets containing chemical structures and associated properties. However, there are currently no algorithms to visualize such data while preserving both global and local features with a sufficient level of detail to allow for human inspection and interpretation. Here, we propose a solution to this problem with a new data visualization method, TMAP, capable of representing data sets of up to millions of data points and arbitrary high dimensionality as a two-dimensional tree (http://tmap.gdb.tools). Visualizations based on TMAP are better suited than t-SNE or UMAP for the exploration and interpretation of large data sets due to their tree-like nature, increased local and global neighborhood and structure preservation, and the transparency of the methods the algorithm is based on. We apply TMAP to the most used chemistry data sets including databases of molecules such as ChEMBL, FDB17, the Natural Products Atlas, DSSTox, as well as to the MoleculeNet benchmark collection of data sets. We also show its broad applicability with further examples from biology, particle physics, and literature.

ChEMBL-Likeness Score and Database GDBChEMBL

The generated database GDB17 enumerates 166.4 billion molecules up to 17 atoms of C, N, O, S and halogens following simple rules of chemical stability and synthetic feasibility. However, most molecules in GDB17 are too complex to be considered for chemical synthesis. To address this limitation, we report GDBChEMBL as a subset of GDB17 featuring 10 million molecules selected according to a ChEMBL-likeness score (CLscore) calculated from the frequency of occurrence of circular substructures in ChEMBL, followed by uniform sampling across molecular size, stereocenters and heteroatoms. Compared to the previously reported subsets FDB17 and GDBMedChem selected from GDB17 by fragment-likeness, respectively, medicinal chemistry criteria, our new subset features molecules with higher synthetic accessibility and possibly bioactivity yet retains a broad and continuous coverage of chemical space typical of the entire GDB17. GDBChEMBL is accessible at http://gdb.unibe.ch for download and for browsing using an interactive chemical space map at http://faerun.gdb.tools.

Freely Accessible Chemical Database Resources of Compounds for In Silico Drug Discovery

BACKGROUND

In silico drug discovery has been proved to be a solidly established key component in early drug discovery. However, this task is hampered by the limitation of quantity and quality of compound databases for screening. In order to overcome these obstacles, freely accessible database resources of compounds have bloomed in recent years. Nevertheless, how to choose appropriate tools to treat these freely accessible databases is crucial. To the best of our knowledge, this is the first systematic review on this issue.

OBJECTIVE

The existed advantages and drawbacks of chemical databases were analyzed and summarized based on the collected six categories of freely accessible chemical databases from literature in this review.

RESULTS

Suggestions on how and in which conditions the usage of these databases could be reasonable were provided. Tools and procedures for building 3D structure chemical libraries were also introduced.

CONCLUSION

In this review, we described the freely accessible chemical database resources for in silico drug discovery. In particular, the chemical information for building chemical database appears as attractive resources for drug design to alleviate experimental pressure.

Generative topographic mapping in drug design

This is a review article of Generative Topographic Mapping (GTM) - a non-linear dimensionality reduction technique producing generative 2D maps of high-dimensional vector spaces - and its specific applications in Drug Design (chemical space cartography, compound library design and analysis, virtual screening, pharmacological profiling, de novo drug design, conformational space & docking interaction cartography, etc.) Written by chemoinformaticians for potential users among medicinal chemists and biologists, the article purposely avoids all underlying mathematics. First, the GTM concept is intuitively explained, based on the strong analogies with the rather popular Self-Organizing Maps (SOMs), which are well established library analysis tools. GTM is basically a fuzzy-logics-based generalization of SOMs. The second part of the review, some of published GTM applications in drug design are briefly revisited.

Energy-Geometry Dependency of Molecular Structures: A Multistep Machine Learning Approach

There is growing interest in estimating quantum observables while circumventing expensive computational overhead for facile in silico materials screening. Machine learning (ML) methods are implemented to perform such calculations in shorter times. Here, we introduce a multistep method based on machine learning algorithms to estimate total energy on the basis of spatial coordinates and charges for various chemical structures, including organic molecules, inorganic molecules, and ions. This method quickly calculates total energy with 0.76 au in root-mean-square error (RMSE) and 1.5% in mean absolute percent error (MAPE) when tested on a database of optimized and unoptimized structures. Using similar molecular representations, experimental thermochemical properties were estimated, with MAPE as low as 6% and RMSE of 8 cal/mol·K for heat capacity in a 10-fold cross-validation.

Method for Systematic Analogue Search Using the Mega SAR Matrix Database

Analogue searching is a typical requirement in hit expansion, hit-to-lead, and lead optimization projects. A new computational methodology is introduced to search for existing and virtual analogues of active compounds. The approach is based upon the SAR matrix (SARM) data structure that was originally developed for the systematic identification and structural organization of analogue series. The SARM-based analogue search algorithm further extends the capacity of current substructure-based methods by (i) simultaneously considering existing and virtual analogues that populate chemical space around query compounds, (ii) permitting not only R-group replacements but also well-defined chemical modifications in core structures to further expand the analogue space, and (iii) automatically extracting all possible analogues from large pools. In addition, as a basis for analogue searching following the SARM concept, the Mega-SARM database is introduced. Mega-SARM is derived from nearly 3.7 million compounds and contains ∼250 000 matrices with structurally related analogue series and more than 1.5 million virtual candidate compounds.

How Size Matters: Diversity for Fragment Library Design

Fragment-based drug discovery (FBDD) has become a major strategy to derive novel lead candidates for various therapeutic targets, as it promises efficient exploration of chemical space by employing fragment-sized (MW < 300) compounds. One of the first challenges in implementing a FBDD approach is the design of a fragment library, and more specifically, the choice of its size and individual members. A diverse set of fragments is required to maximize the chances of discovering novel hit compounds. However, the exact diversity of a certain collection of fragments remains underdefined, which hinders direct comparisons among different selections of fragments. Based on structural fingerprints, we herein introduced quantitative metrics for the structural diversity of fragment libraries. Structures of commercially available fragments were retrieved from the ZINC database, from which libraries with sizes ranging from 100 to 100,000 compounds were selected. The selected libraries were evaluated and compared quantitatively, resulting in interesting size-diversity relationships. Our results demonstrated that while library size does matter for its diversity, there exists an optimal size for structural diversity. It is also suggested that such quantitative measures can guide the design of diverse fragment libraries under different circumstances.

Concepts of Artificial Intelligence for Computer-Assisted Drug Discovery

Artificial intelligence (AI), and, in particular, deep learning as a subcategory of AI, provides opportunities for the discovery and development of innovative drugs. Various machine learning approaches have recently (re)emerged, some of which may be considered instances of domain-specific AI which have been successfully employed for drug discovery and design. This review provides a comprehensive portrayal of these machine learning techniques and of their applications in medicinal chemistry. After introducing the basic principles, alongside some application notes, of the various machine learning algorithms, the current state-of-the art of AI-assisted pharmaceutical discovery is discussed, including applications in structure- and ligand-based virtual screening, de novo drug design, physicochemical and pharmacokinetic property prediction, drug repurposing, and related aspects. Finally, several challenges and limitations of the current methods are summarized, with a view to potential future directions for AI-assisted drug discovery and design.

Medicinal Chemistry Aware Database GDBMedChem

The generated database GDB17 enumerates 166.4 billion possible molecules up to 17 atoms of C, N, O, S and halogens following simple chemical stability and synthetic feasibility rules, however medicinal chemistry criteria are not taken into account. Here we applied rules inspired by medicinal chemistry to exclude problematic functional groups and complex molecules from GDB17, and sampled the resulting subset uniformly across molecular size, stereochemistry and polarity to form GDBMedChem as a compact collection of 10 million small molecules. This collection has reduced complexity and better synthetic accessibility than the entire GDB17 but retains higher sp³ -carbon fraction and natural product likeness scores compared to known drugs. GDBMedChem molecules are more diverse and very different from known molecules in terms of substructures and represent an unprecedented source of diversity for drug design. GDBMedChem is available for 3D-visualization, similarity searching and for download at http://gdb.unibe.ch.

Design of a Fragment-Screening Library

Herein we describe a method for the design, purchase, and assembly of a fragment-screening library from a list of commercially available compounds. The computational tools used in assessment of compound properties as well as the workflow for compound selection are provided for reference as implemented in commercially available software that is free and accessible to most academic users. The workflow can be modified as necessary to generate a fit-for-purpose fragment library with the desired compound property profiles. An analytical process for assessing the quality, identity, and suitability of a purchased fragment for inclusion in a screening collection is described. Results from our in-house library are presented as an example of compound progression through this quality control process.

Selectivity Challenges in Docking Screens for GPCR Targets and Antitargets

To investigate large library docking's ability to find molecules with joint activity against on-targets and selectivity versus antitargets, the dopamine D₂ and serotonin 5-HT_2A receptors were targeted, seeking selectivity against the histamine H₁ receptor. In a second campaign, κ-opioid receptor ligands were sought with selectivity versus the μ-opioid receptor. While hit rates ranged from 40% to 63% against the on-targets, they were just as good against the antitargets, even though the molecules were selected for their putative lack of binding to the off-targets. Affinities, too, were often as good or better for the off-targets. Even though it was occasionally possible to find selective molecules, such as a mid-nanomolar D₂/5-HT_2A ligand with 21-fold selectivity versus the H₁ receptor, this was the exception. Whereas false-negatives are tolerable in docking screens against on-targets, they are intolerable against antitargets; addressing this problem may demand new strategies in the field.

Efficiently Exploring Adsorption Space to Identify Privileged Adsorbents for Chemical Separations of a Diverse Set of Molecules

Although computational models have been used to predict adsorption of molecules in large libraries of porous adsorbents, previous work of this kind has focused on a small number of molecules as potential adsorbates. In this study, molecular simulations were used to consider the adsorption of a diverse range of molecules in a large collection of metal-organic framework (MOF) materials. Specifically, 11 304 isotherms were obtained from molecular simulations of 24 different adsorbates in 471 MOFs. This information provides insight into several interesting questions that could not be addressed with previously available data. Highly computationally efficient methods are introduced that can predict isotherms for a wide range of adsorbing molecules with far less computation than traditional molecular simulations. By characterizing the 276 binary mixtures defined by the molecules considered, "privileged" adsorbents are shown to exist, which are effective for separating many different molecular mixtures. Finally, correlations that were developed previously to predict molecular solubility in polymers are found to be surprisingly effective in predicting the average properties of molecules adsorbing in MOFs.

The Alexandria library, a quantum-chemical database of molecular properties for force field development

Data quality as well as library size are crucial issues for force field development. In order to predict molecular properties in a large chemical space, the foundation to build force fields on needs to encompass a large variety of chemical compounds. The tabulated molecular physicochemical properties also need to be accurate. Due to the limited transparency in data used for development of existing force fields it is hard to establish data quality and reusability is low. This paper presents the Alexandria library as an open and freely accessible database of optimized molecular geometries, frequencies, electrostatic moments up to the hexadecupole, electrostatic potential, polarizabilities, and thermochemistry, obtained from quantum chemistry calculations for 2704 compounds. Values are tabulated and where available compared to experimental data. This library can assist systematic development and training of empirical force fields for a broad range of molecules.

Big Data in Drug Discovery

Interpretation of Big Data in the drug discovery community should enhance project timelines and reduce clinical attrition through improved early decision making. The issues we encounter start with the sheer volume of data and how we first ingest it before building an infrastructure to house it to make use of the data in an efficient and productive way. There are many problems associated with the data itself including general reproducibility, but often, it is the context surrounding an experiment that is critical to success. Help, in the form of artificial intelligence (AI), is required to understand and translate the context. On the back of natural language processing pipelines, AI is also used to prospectively generate new hypotheses by linking data together. We explain Big Data from the context of biology, chemistry and clinical trials, showcasing some of the impressive public domain sources and initiatives now available for interrogation.

Mapping of the Available Chemical Space versus the Chemical Universe of Lead-Like Compounds

This is, to our knowledge, the most comprehensive analysis to date based on generative topographic mapping (GTM) of fragment-like chemical space (40 million molecules with no more than 17 heavy atoms, both from the theoretically enumerated GDB-17 and real-world PubChem/ChEMBL databases). The challenge was to prove that a robust map of fragment-like chemical space can actually be built, in spite of a limited (≪10⁵ ) maximal number of compounds ("frame set") usable for fitting the GTM manifold. An evolutionary map building strategy has been updated with a "coverage check" step, which discards manifolds failing to accommodate compounds outside the frame set. The evolved map has a good propensity to separate actives from inactives for more than 20 external structure-activity sets. It was proven to properly accommodate the entire collection of 40 m compounds. Next, it served as a library comparison tool to highlight biases of real-world molecules (PubChem and ChEMBL) versus the universe of all possible species represented by FDB-17, a fragment-like subset of GDB-17 containing 10 million molecules. Specific patterns, proper to some libraries and absent from others (diversity holes), were highlighted.

SmilesDrawer: Parsing and Drawing SMILES-Encoded Molecular Structures Using Client-Side JavaScript

Here we present SmilesDrawer, a dependency-free JavaScript component capable of both parsing and drawing SMILES-encoded molecular structures client-side, developed to be easily integrated into web projects and to display organic molecules in large numbers and fast succession. SmilesDrawer can draw structurally and stereochemically complex structures such as maitotoxin and C₆₀ without using templates, yet has an exceptionally small computational footprint and low memory usage without the requirement for loading images or any other form of client-server communication, making it easy to integrate even in secure (intranet, firewalled) or offline applications. These features allow the rendering of thousands of molecular structure drawings on a single web page within seconds on a wide range of hardware supporting modern browsers. The source code as well as the most recent build of SmilesDrawer is available on Github ( http://doc.gdb.tools/smilesDrawer/ ). Both yarn and npm packages are also available.

Analysing and Navigating Natural Products Space for Generating Small, Diverse, But Representative Chemical Libraries

Armed with the digital availability of two natural products libraries, amounting to some 195 885 molecular entities, we ask the question of how we can best sample from them to maximize their "representativeness" in smaller and more usable libraries of 96, 384, 1152, and 1920 molecules. The term "representativeness" is intended to include diversity, but for numerical reasons (and the likelihood of being able to perform a QSAR) it is necessary to focus on areas of chemical space that are more highly populated. Encoding chemical structures as fingerprints using the RDKit "patterned" algorithm, we first assess the granularity of the natural products space using a simple clustering algorithm, showing that there are major regions of "denseness" but also a great many very sparsely populated areas. We then apply a "hybrid" hierarchical K-means clustering algorithm to the data to produce more statistically robust clusters from which representative and appropriate numbers of samples may be chosen. There is necessarily again a trade-off between cluster size and cluster number, but within these constraints, libraries containing 384 or 1152 molecules can be found that come from clusters that represent some 18 and 30% of the whole chemical space, with cluster sizes of, respectively, 50 and 27 or above, just about sufficient to perform a QSAR. By using the online availability of molecules via the Molport system (www.molport.com), we are also able to construct (and, for the first time, provide the contents of) a small virtual library of available molecules that provided effective coverage of the chemical space described. Consistent with this, the average molecular similarities of the contents of the libraries developed is considerably smaller than is that of the original libraries. The suggested libraries may have use in molecular or phenotypic screening, including for determining possible transporter substrates.

Virtual Exploration of the Ring Systems Chemical Universe

Here, we explore the chemical space of all virtually possible organic molecules focusing on ring systems, which represent the cyclic cores of organic molecules obtained by removing all acyclic bonds and converting all remaining atoms to carbon. This approach circumvents the combinatorial explosion encountered when enumerating the molecules themselves. We report the chemical universe database GDB4c containing 916 130 ring systems up to four saturated or aromatic rings and maximum ring size of 14 atoms and GDB4c3D containing the corresponding 6 555 929 stereoisomers. Almost all (98.6%) of these ring systems are unknown and represent chiral 3D-shaped macrocycles containing small rings and quaternary centers reminiscent of polycyclic natural products. We envision that GDB4c can serve to select new ring systems from which to design analogs of such natural products. The database is available for download at www.gdb.unibe.ch together with interactive visualization and search tools as a resource for molecular design.