Artificial intelligence in chemistry and drug design

From Pipeline to Plant Protection Products: Using New Approach Methodologies (NAMs) in Agrochemical Safety Assessment

The human population will be approximately 9.7 billion by 2050, and food security has been identified as one of the key issues facing the global population. Agrochemicals are an important tool available to farmers that enable high crop yields and continued access to healthy foods, but the average new agrochemical active ingredient takes more than ten years, 350 million dollars, and 20,000 animals to develop and register. The time, monetary, and animal costs incentivize the use of New Approach Methodologies (NAMs) in early-stage screening to prioritize chemical candidates. This review outlines NAMs that are currently available or can be adapted for use in early-stage screening agrochemical programs. It covers new in vitro screens that are on the horizon in key areas of regulatory concern. Overall, early-stage screening with NAMs enables the prioritization of development for agrochemicals without human and environmental health concerns through a more directed, agile, and iterative development program before animal-based regulatory testing is even considered.

Artifical intelligence: a virtual chemist for natural product drug discovery

Nature is full of a bundle of medicinal substances and its product perceived as a prerogative structure to collaborate with protein drug targets. The natural product's (NPs) structure heterogeneity and eccentric characteristics inspired scientists to work on natural product-inspired medicine. To gear NP drug-finding artificial intelligence (AI) to confront and excavate unexplored opportunities. Natural product-inspired drug discoveries based on AI to act as an innovative tool for molecular design and lead discovery. Various models of machine learning produce quickly synthesizable mimetics of the natural products templates. The invention of novel natural products mimetics by computer-assisted technology provides a feasible strategy to get the natural product with defined bio-activities. AI's hit rate makes its high importance by improving trail patterns such as dose selection, trail life span, efficacy parameters, and biomarkers. Along these lines, AI methods can be a successful tool in a targeted way to formulate advanced medicinal applications for natural products. 'Prediction of future of natural product based drug discovery is not magic, actually its artificial intelligence'Communicated by Ramaswamy H. Sarma.

Knowledge, attitude and practice among pharmacy students and faculty members towards artificial intelligence in pharmacy practice: A multinational cross-sectional study

BACKGROUND

Modern patient care depends on the continuous improvement of community and clinical pharmacy services, and artificial intelligence (AI) has the potential to play a key role in this evolution. Although AI has been increasingly implemented in various fields of pharmacy, little is known about the knowledge, attitudes, and practices (KAP) of pharmacy students and faculty members towards this technology.

OBJECTIVES

The primary objective of this study was to investigate the KAP of pharmacy students and faculty members regarding AI in six countries in the Middle East as well as to identify the predictive factors behind the understanding of the principles and practical applications of AI in healthcare processes.

MATERIAL AND METHODS

This study was a descriptive cross-sectional survey. A total of 875 pharmacy students and faculty members in the faculty of pharmacy in Jordan, Palestine, Lebanon, Egypt, Saudi Arabia, and Libya participated in the study. Data was collected through an online electronic questionnaire. The data collected included information about socio-demographics, understanding of AI basic principles, participants' attitudes toward AI, the participants' AI practices.

RESULTS

Most participants (92.6%) reported having heard of AI technology in their practice, but only a small proportion (39.5%) had a good understanding of its concepts. The overall level of knowledge about AI among the study participants was moderate, with the mean knowledge score being 42.3 ± 21.8 out of 100 and students having a significantly higher knowledge score than faculty members. The attitude towards AI among pharmacy students and faculty members was positive, but there were still concerns about the impact of AI on job security and patient safety. Pharmacy students and faculty members had limited experience using AI tools in their practice. The majority of respondents (96.2%) believed that AI could improve patient care and pharmacy services. However, only a minority (18.6%) reported having received education or training on AI technology. High income, a strong educational level and background, and previous experience with technologies were predictors of KAP toward using AI in pharmacy practice. Finally, there was a positive correlation between knowledge about AI and attitudes towards AI as well as a significant positive correlation between AI knowledge and overall KAP scores.

CONCLUSION

The findings suggest that while there is a growing awareness of AI technology among pharmacy professionals in the Middle East and North Africa (MENA) region, there are still significant gaps in understanding and adopting AI in pharmacy Practice.

Characterizing emerging companies in computational drug development

Computation promises to accelerate, de-risk and optimize drug research and development. An increasing number of companies have entered this space, specializing in the design of new algorithms, computing on proprietary data, and/or development of hardware to improve distinct drug pipeline stages. The large number of such companies and their unique strategies and deals have created a highly complex and competitive industry. We comprehensively analyze the companies in this space to highlight trends and opportunities, identifying highly occupied areas of risk and currently underrepresented niches of high value.

The Application of Artificial Intelligence and Drug Repositioning for the Identification of Fibroblast Growth Factor Receptor Inhibitors: A Review

Artificial intelligence talks about modeling intelligent behavior through a computer with the least human involvement. Drug repositioning techniques based on artificial intelligence accelerate the research process and decrease the cost of experimental studies. Dysregulation of fibroblast growth factor (FGF) receptors as the tyrosine kinase family of receptors plays a vital role in a wide range of malignancies. Because of their functional significance, they were considered promising drug targets for the therapy of various cancers. This review has summarized small molecules capable of inhibiting FGF receptors that progressed using artificial intelligence and repositioning drugs examined in clinical trials associated with cancer therapy. This review is based on a literature search in PubMed, Web of Science, Scopus EMBASE, and Google Scholar databases to gather the necessary information in each chapter by employing keywords like artificial intelligence, computational drug design, drug repositioning, and FGF receptor inhibitors. To achieve this goal, a spacious literature review of human studies in these fields-published over the last 20 decades-was performed. According to published reports, nonselective FGF receptor inhibitors can be used for cancer management, and multitarget kinase inhibitors are the first drug class approved due to more advanced clinical studies. For example, AZD4547 and BGJ398 are gradually entering the consumption cycle and are good options as combined treatments. Artificial intelligence and drug repositioning methods can help preselect suitable drug targets more successfully for future inhibition of carcinogenicity.

Photochemically Activated 3D Printing Inks: Current Status, Challenges, and Opportunities

3D printing with light is enabled by the photochemistry underpinning it. Without fine control over the ability to photochemically gate covalent bond formation by the light at a certain wavelength and intensity, advanced photoresists with functions spanning from on-demand degradability, adaptability, rapid printing speeds, and tailored functionality are impossible to design. Herein, recent advances in photoresist design for light-driven 3D printing applications are critically assessed, and an outlook of the outstanding challenges and opportunities is provided. This is achieved by classing the discussed photoresists in chemistries that function photoinitiator-free and those that require a photoinitiator to proceed. Such a taxonomy is based on the efficiency with which photons are able to generate covalent bonds, with each concept featuring distinct advantages and drawbacks.

n-Tuples on Scaffold Diversity Inspired by Drug Hybridisation to Enhance Drugability: Application to Cytarabine

A mathematical concept, n-tuples are originally applied to medicinal chemistry, especially with the creation of scaffold diversity inspired by the hybridisation of different commercial drugs with cytarabine, a synthetic arabinonucleoside derived from two marine natural products, spongouridine and spongothymidine. The new methodology explores the virtual chemical-factorial combination of different commercial drugs (immunosuppressant, antibiotic, antiemetic, anti-inflammatory, and anticancer) with the anticancer drug cytarabine. Real chemical combinations were designed and synthesised for 8-duples, obtaining a small representative library of interesting organic molecules to be biologically tested as proof of concept. The synthesised library contains classical molecular properties regarding the Lipinski rules and/or beyond rules of five (bRo5) and is represented by the covalent combination of the anticancer drug cytarabine with ibuprofen, flurbiprofen, folic acid, sulfasalazine, ciprofloxacin, bortezomib, and methotrexate. The insertion of specific nomenclature could be implemented into artificial intelligence algorithms in order to enhance the efficiency of drug-hunting programs. The novel methodology has proven useful for the straightforward synthesis of most of the theoretically proposed duples and, in principle, could be extended to any other central drug.

DeepDelta: predicting ADMET improvements of molecular derivatives with deep learning

Established molecular machine learning models process individual molecules as inputs to predict their biological, chemical, or physical properties. However, such algorithms require large datasets and have not been optimized to predict property differences between molecules, limiting their ability to learn from smaller datasets and to directly compare the anticipated properties of two molecules. Many drug and material development tasks would benefit from an algorithm that can directly compare two molecules to guide molecular optimization and prioritization, especially for tasks with limited available data. Here, we develop DeepDelta, a pairwise deep learning approach that processes two molecules simultaneously and learns to predict property differences between two molecules from small datasets. On 10 ADMET benchmark tasks, our DeepDelta approach significantly outperforms two established molecular machine learning algorithms, the directed message passing neural network (D-MPNN) ChemProp and Random Forest using radial fingerprints, for 70% of benchmarks in terms of Pearson's r, 60% of benchmarks in terms of mean absolute error (MAE), and all external test sets for both Pearson's r and MAE. We further analyze our performance and find that DeepDelta is particularly outperforming established approaches at predicting large differences in molecular properties and can perform scaffold hopping. Furthermore, we derive mathematically fundamental computational tests of our models based on mathematical invariants and show that compliance to these tests correlates with overall model performance - providing an innovative, unsupervised, and easily computable measure of expected model performance and applicability. Taken together, DeepDelta provides an accurate approach to predict molecular property differences by directly training on molecular pairs and their property differences to further support fidelity and transparency in molecular optimization for drug development and the chemical sciences.

Bio-inspired microfluidics: A review

Biomicrofluidics, a subdomain of microfluidics, has been inspired by several ideas from nature. However, while the basic inspiration for the same may be drawn from the living world, the translation of all relevant essential functionalities to an artificially engineered framework does not remain trivial. Here, we review the recent progress in bio-inspired microfluidic systems via harnessing the integration of experimental and simulation tools delving into the interface of engineering and biology. Development of "on-chip" technologies as well as their multifarious applications is subsequently discussed, accompanying the relevant advancements in materials and fabrication technology. Pointers toward new directions in research, including an amalgamated fusion of data-driven modeling (such as artificial intelligence and machine learning) and physics-based paradigm, to come up with a human physiological replica on a synthetic bio-chip with due accounting of personalized features, are suggested. These are likely to facilitate physiologically replicating disease modeling on an artificially engineered biochip as well as advance drug development and screening in an expedited route with the minimization of animal and human trials.

Interpretable Molecular Property Predictions Using Marginalized Graph Kernels

Marginalized graph kernels have shown competitive performance in molecular machine learning tasks but currently lack measures of interpretability, which are important to improve trust in the models, detect biases, and inform molecular optimization campaigns. We here conceive and implement two interpretability measures for Gaussian process regression using a marginalized graph kernel (GPR-MGK) to quantify (1) the contribution of specific training data to the prediction and (2) the contribution of specific nodes of the graph to the prediction. We demonstrate the applicability of these interpretability measures for molecular property prediction. We compare GPR-MGK to graph neural networks on four logic and two real-world toxicology data sets and find that the atomic attribution of GPR-MGK generally outperforms the atomic attribution of graph neural networks. We also perform a detailed molecular attribution analysis using the FreeSolv data set, showing how molecules in the training set influence machine learning predictions and why Morgan fingerprints perform poorly on this data set. This is the first systematic examination of the interpretability of GPR-MGK and thereby is an important step in the further maturation of marginalized graph kernel methods for interpretable molecular predictions.

Molecular Representations in Machine-Learning-Based Prediction of PK Parameters for Insulin Analogs

Therapeutic peptides and proteins derived from either endogenous hormones, such as insulin, or de novo design via display technologies occupy a distinct pharmaceutical space in between small molecules and large proteins such as antibodies. Optimizing the pharmacokinetic (PK) profile of drug candidates is of high importance when it comes to prioritizing lead candidates, and machine-learning models can provide a relevant tool to accelerate the drug design process. Predicting PK parameters of proteins remains difficult due to the complex factors that influence PK properties; furthermore, the data sets are small compared to the variety of compounds in the protein space. This study describes a novel combination of molecular descriptors for proteins such as insulin analogs, where many contained chemical modifications, e.g., attached small molecules for protraction of the half-life. The underlying data set consisted of 640 structural diverse insulin analogs, of which around half had attached small molecules. Other analogs were conjugated to peptides, amino acid extensions, or fragment crystallizable regions. The PK parameters clearance (CL), half-life (T1/2), and mean residence time (MRT) could be predicted by using classical machine-learning models such as Random Forest (RF) and Artificial Neural Networks (ANN) with root-mean-square errors of CL of 0.60 and 0.68 (log units) and average fold errors of 2.5 and 2.9 for RF and ANN, respectively. Both random and temporal data splittings were employed to evaluate ideal and prospective model performance with the best models, regardless of data splitting, achieving a minimum of 70% of predictions within a twofold error. The tested molecular representations include (1) global physiochemical descriptors combined with descriptors encoding the amino acid composition of the insulin analogs, (2) physiochemical descriptors of the attached small molecule, (3) protein language model (evolutionary scale modeling) embedding of the amino acid sequence of the molecules, and (4) a natural language processing inspired embedding (mol2vec) of the attached small molecule. Encoding the attached small molecule via (2) or (4) significantly improved the predictions, while the benefit of using the protein language model-based encoding (3) depended on the used machine-learning model. The most important molecular descriptors were identified as descriptors related to the molecular size of both the protein and protraction part using Shapley additive explanations values. Overall, the results show that combining representations of proteins and small molecules was key for PK predictions of insulin analogs.

67 million natural product-like compound database generated via molecular language processing

Natural products are a rich resource of bioactive compounds for valuable applications across multiple fields such as food, agriculture, and medicine. For natural product discovery, high throughput in silico screening offers a cost-effective alternative to traditional resource-heavy assay-guided exploration of structurally novel chemical space. In this data descriptor, we report a characterized database of 67,064,204 natural product-like molecules generated using a recurrent neural network trained on known natural products, demonstrating a significant 165-fold expansion in library size over the approximately 400,000 known natural products. This study highlights the potential of using deep generative models to explore novel natural product chemical space for high throughput in silico discovery.

Graph Polish: A Novel Graph Generation Paradigm for Molecular Optimization

Molecular optimization, which transforms a given input molecule X into another Y with desired properties, is essential in molecular drug discovery. The traditional approaches either suffer from sample-inefficient learning or ignore information that can be captured with the supervised learning of optimized molecule pairs. In this study, we present a novel molecular optimization paradigm, Graph Polish. In this paradigm, with the guidance of the source and target molecule pairs of the desired properties, a heuristic optimization solution can be derived: given an input molecule, we first predict which atom can be viewed as the optimization center, and then the nearby regions are optimized around this center. We then propose an effective and efficient learning framework, Teacher and Student polish, to capture the dependencies in the optimization steps. A teacher component automatically identifies and annotates the optimization centers and the preservation, removal, and addition of some parts of the molecules; a student component learns these knowledges and applies them to a new molecule. The proposed paradigm can offer an intuitive interpretation for the molecular optimization result. Experiments with multiple optimization tasks are conducted on several benchmark datasets. The proposed approach achieves a significant advantage over the six state-of-the-art baseline methods. Also, extensive studies are conducted to validate the effectiveness, explainability, and time savings of the novel optimization paradigm.

Can We Quickly Learn to "Translate" Bioactive Molecules with Transformer Models?

Meaningful exploration of the chemical space of druglike molecules in drug design is a highly challenging task due to a combinatorial explosion of possible modifications of molecules. In this work, we address this problem with transformer models, a type of machine learning (ML) model originally developed for machine translation. By training transformer models on pairs of similar bioactive molecules from the public ChEMBL data set, we enable them to learn medicinal-chemistry-meaningful, context-dependent transformations of molecules, including those absent from the training set. By retrospective analysis on the performance of transformer models on ChEMBL subsets of ligands binding to COX2, DRD2, or HERG protein targets, we demonstrate that the models can generate structures identical or highly similar to most active ligands, despite the models having not seen any ligands active against the corresponding protein target during training. Our work demonstrates that human experts working on hit expansion in drug design can easily and quickly employ transformer models, originally developed to translate texts from one natural language to another, to "translate" from known molecules active against a given protein target to novel molecules active against the same target.

New avenues in artificial-intelligence-assisted drug discovery

Over the past decade, the amount of biomedical data available has grown at unprecedented rates. Increased automation technology and larger data volumes have encouraged the use of machine learning (ML) or artificial intelligence (AI) techniques for mining such data and extracting useful patterns. Because the identification of chemical entities with desired biological activity is a crucial task in drug discovery, AI technologies have the potential to accelerate this process and support decision making. In addition, the advent of deep learning (DL) has shown great promise in addressing diverse problems in drug discovery, such as de novo molecular design. Herein, we will appraise the current state-of-the-art in AI-assisted drug discovery, discussing the recent applications covering generative models for chemical structure generation, scoring functions to improve binding affinity and pose prediction, and molecular dynamics to assist in the parametrization, featurization and generalization tasks. Finally, we will discuss current hurdles and the strategies to overcome them, as well as potential future directions.

Computational Approaches to Designing Antiviral Drugs against COVID-19: A Comprehensive Review

The global impact of the COVID-19 pandemic caused by SARS-CoV-2 necessitates innovative strategies for the rapid development of effective treatments. Computational methodologies, such as molecular modelling, molecular dynamics simulations, and artificial intelligence, have emerged as indispensable tools in the drug discovery process. This review aimed to provide a comprehensive overview of these computational approaches and their application in the design of antiviral agents for COVID-19. Starting with an examination of ligand-based and structure-based drug discovery, the review has delved into the intricate ways through which molecular modelling can accelerate the identification of potential therapies. Additionally, the investigation extends to phytochemicals sourced from nature, which have shown promise as potential antiviral agents. Noteworthy compounds, including gallic acid, naringin, hesperidin, Tinospora cordifolia, curcumin, nimbin, azadironic acid, nimbionone, nimbionol, and nimocinol, have exhibited high affinity for COVID-19 Mpro and favourable binding energy profiles compared to current drugs. Although these compounds hold potential, their further validation through in vitro and in vivo experimentation is imperative. Throughout this exploration, the review has emphasized the pivotal role of computational biologists, bioinformaticians, and biotechnologists in driving rapid advancements in clinical research and therapeutic development. By combining state-of-the-art computational techniques with insights from structural and molecular biology, the search for potent antiviral agents has been accelerated. The collaboration between these disciplines holds immense promise in addressing the transmissibility and virulence of SARS-CoV-2.

To Affinity and Beyond: A Personal Reflection on the Design and Discovery of Drugs

Faced with new and as yet unmet medical need, the stark underperformance of the pharmaceutical discovery process is well described if not perfectly understood. Driven primarily by profit rather than societal need, the search for new pharmaceutical products—small molecule drugs, biologicals, and vaccines—is neither properly funded nor sufficiently systematic. Many innovative approaches remain significantly underused and severely underappreciated, while dominant methodologies are replete with problems and limitations. Design is a component of drug discovery that is much discussed but seldom realised. In and of itself, technical innovation alone is unlikely to fulfil all the possibilities of drug discovery if the necessary underlying infrastructure remains unaltered. A fundamental revision in attitudes, with greater reliance on design powered by computational approaches, as well as a move away from the commercial imperative, is thus essential to capitalise fully on the potential of pharmaceutical intervention in healthcare.

The New Era of Drug Discovery: The Power of Computer-aided Drug Design (CADD)

Abstract:

Drug design and discovery is a process that requires high financial costs and is timeconsuming. For many years, this process focused on empirical pharmacology. However, over the years, the target-based approach allowed a significant discovery in this field, initiating the rational design era. In view, to decrease the time and financial cost, rational drug design is benefited by increasing computer engineering and software development, and computer-aided drug design (CADD) emerges as a promising alternative. Since the 1970s, this approach has been able to identify many important and revolutionary compounds, like protease inhibitors, antibiotics, and others. Many anticancer compounds identified through this approach have shown their importance, being CADD essential in any drug discovery campaign. Thus, this perspective will present the prominent successful cases utilizing this approach and entering into the next stage of drug design. We believe that drug discovery will follow the progress in bioinformatics, using high-performance computing with molecular dynamics protocols faster and more effectively. In addition, artificial intelligence and machine learning will be the next process in the rational design of new drugs. Here, we hope that this paper generates new ideas and instigates research groups worldwide to use these methods and stimulate progress in drug design.

Molecular Screening and Toxicity Estimation of 260,000 Perfluoroalkyl and Polyfluoroalkyl Substances (PFASs) through Machine Learning

Perfluoroalkyl and polyfluoroalkyl substances (PFASs) are a class of chemicals widely used in industrial applications due to their exceptional properties and stability. However, they do not readily degrade in the environment and are linked to contamination and adverse health effects in humans and wildlife. To find alternatives for the most commonly used PFAS molecules that maintain their desirable chemical properties but are not adverse to biological lifeforms, a novel approach based upon machine learning is utilized. The machine learning model is trained on an existing set of PFAS molecules to generate over 260,000 novel PFAS molecules, which we dub PFAS-AI-Gen. Using molecular descriptors with known relationships to toxicity and industrial suitability followed by molecular docking and molecular dynamics simulations, this set of molecules is screened. In this manner, increasingly complex calculations are performed only for candidate molecules that are most likely to yield the desired properties of low binding affinity toward two selected protein receptors, the human pregnane x receptor (hPXR) and peroxisome proliferator-activated receptor γ (PPAR-γ), and high industrial suitability, defined by critical micelle concentration (CMC). The selection criteria of low binding affinity and high industrial suitability are relative to the popular PFAS alternative GenX. hPXR and PPAR-γ are selected as they are PFAS targets and facilitate a variety of functions, such as drug metabolism and glucose regulation, respectively. Through this approach, 22 promising new PFAS substitutes that may warrant experimental investigation are identified. This integrated approach of molecular screening and toxicity estimation may be applicable to other chemical classes.

Effect of gem-Difluorination on the Key Physicochemical Properties Relevant to Medicinal Chemistry: The Case of Functionalized Cycloalkanes

Physico-chemical properties important to drug discovery (pK_a , LogP, and aqueous solubility), as well as metabolic stability, were studied for a series of functionalized gem-difluorinated cycloalkanes and compared to those of non-fluorinated and acyclic counterparts to evaluate the impact of the fluorination. It was found that the influence of the CF₂ moiety on the acidity/basicity of the corresponding carboxylic acids and amines was defined by inductive the effect of the fluorine atoms and was nearly the same for acyclic and cyclic aliphatic compounds. Lipophilicity and aqueous solubility followed more complex trends and were affected by the position of the fluorine atoms, ring size, and even the nature of the functional group present; also, significant differences were found for the acyclic and cyclic series. Also, gem-difluorination either did not affect or slightly improved the metabolic stability of the corresponding model derivatives. The presented results can be used as a guide for rational drug design employing fluorine and establish the first chapter in a catalog of the key in vitro properties of fluorinated cycloalkanes.

Natural product drug discovery in the artificial intelligence era

Natural products (NPs) are primarily recognized as privileged structures to interact with protein drug targets. Their unique characteristics and structural diversity continue to marvel scientists for developing NP-inspired medicines, even though the pharmaceutical industry has largely given up. High-performance computer hardware, extensive storage, accessible software and affordable online education have democratized the use of artificial intelligence (AI) in many sectors and research areas. The last decades have introduced natural language processing and machine learning algorithms, two subfields of AI, to tackle NP drug discovery challenges and open up opportunities. In this article, we review and discuss the rational applications of AI approaches developed to assist in discovering bioactive NPs and capturing the molecular "patterns" of these privileged structures for combinatorial design or target selectivity.

The Brazilian compound library (BraCoLi) database: a repository of chemical and biological information for drug design

The Brazilian Compound Library (BraCoLi) is a novel open access and manually curated electronic library of compounds developed by Brazilian research groups to support further computer-aided drug design works, available on https://www.farmacia.ufmg.br/qf/downloads/ . Herein, the first version of the database is described comprising 1176 compounds. Also, the chemical diversity and drug-like profiles of BraCoLi were defined to analyze its chemical space. A significant amount of the compounds fitted Lipinski and Veber's rules, alongside other drug-likeness properties. A comparison using principal component analysis showed that BraCoLi is similar to other databases (FDA-approved drugs and NuBBE_DB) regarding structural and physicochemical patterns. Furthermore, a scaffold analysis showed that BraCoLi presents several privileged chemical skeletons with great diversity. Despite the similar distribution in the structural and physicochemical spaces, Tanimoto coefficient values indicated that compounds present in the BraCoLi are generally different from the two other databases, where they showed different kernel distributions and low similarity. These facts show an interesting innovative aspect, which is a desirable feature for novel drug design purposes.

Development and Evaluation of Geometry Optimization Algorithms in Conjunction with ANI Potentials

An efficient yet accurate method for producing a large amount of energy data for molecular mechanical force field (MMFF) parameterization is on demand, especially for torsional angle parameters which are typically derived to reproduce ab initio rotational profiles or torsional potential energy surfaces (PESs). Recently, an active learning potential (ANI-1x) for organic molecules which can produce smooth and physically meaningful PESs has been developed. The high efficiency and accuracy make ANI-1x especially attractive for geometry optimization at low cost. To apply the ANI-1x potential in MMFF parameterization, one needs to perform constrained geometry optimization. In this work, we first developed a computational protocol to constrain rotatable torsional angles and other geometric parameters for a molecule whose geometry is described by Cartesian coordinates. The constraint is successfully achieved by force projection for the two conjugated gradient (CG) algorithms. We then conducted large-scale assessments on ANI-1x along with four different optimization algorithms in reproducing DFT energies and geometries for two CG algorithms, CG backtracking line search (CG-BS) and CG Wolfe line search (CG-WS), and two quasi-Newton algorithms, Broyden-Fletcher-Goldfarb-Shanno (BFGS) and low-memory BFGS (L-BFGS). Note that CG-BS is a new algorithm we developed in this work. All four algorithms take the ANI energies and forces to optimize a molecule geometry. Last, we conducted a large-scale assessment of applying ANI-1x in MMFF development in three aspects. First, we performed full optimizations for 100 drug molecules, each consisting of five distinct conformations. The average root-mean-square error (RMSE) between ANI-1x and DFT is about 1.3 kcal/mol, and the root-mean-square displacement (RMSD) of heavy atoms is about 0.35 Å. Second, we generated torsional PESs for 160 organic molecules, and constrained optimizations were performed for up to 18 conformations for each PES. We found that the RMSE of all the conformers is 1.23 kcal/mol. Last, we carried out constrained optimizations for alanine dipeptide with both ϕ and φ angles being frozen. The Ramachandran plots indicate that the two CG algorithms in conjunction with the ANI-1x potential could well reproduce the DFT-optimized geometries and torsional PESs. We concluded that CG-BS and CG-WS are good choices for generating PESs, while CG-WS or BFGS is ideal for performing full geometry optimization. With the continuously increased quality of ANI, it is expected that the computational algorithms and protocols presented in this work will have great applications in improving the quality of an existing small-molecule MMFF.

Radical Scavenging Activity and Pharmacokinetic Properties of Coumarin-Hydroxybenzohydrazide Hybrids

Free radicals often interact with vital proteins, violating their structure and inhibiting their activity. In previous studies, synthesis, characterisation, and the antioxidative properties of the five different coumarin derivatives have been investigated. In the tests of potential toxicity, all compounds exhibited low toxicity with significant antioxidative potential at the same time. In this paper, the radical scavenging activity of the abovementioned coumarin derivatives towards ten different radical species was investigated. It was found that all investigated compounds show good radical scavenging ability, with results that are in correlation with the results published in the previous study. Three additional mechanisms of radical scavenging activity were investigated. It was found that all three mechanisms are thermodynamically plausible and in competition. Interestingly, it was found that products of the Double Hydrogen Atom Transfer (DHAT) mechanism, a biradical species in triplet spin state, are in some cases more stable than singlet spin state analogues. This unexpected trend can be explained by spin delocalisation over the hydrazide bridge and phenolic part of the molecule with a low probability of spin pairing. Besides radical-scavenging activity, the pharmacokinetic and drug-likeness of the coumarin hybrids were investigated. It was found that they exhibit good membrane and skin permeability and potential interactions with P-450 enzymes. Furthermore, it was found that investigated compounds satisfy all criteria of the drug-likeness tests, suggesting they possess a good preference for being used as potential drugs.

Critical assessment of methods of protein structure prediction (CASP)-Round XIV

Critical assessment of structure prediction (CASP) is a community experiment to advance methods of computing three-dimensional protein structure from amino acid sequence. Core components are rigorous blind testing of methods and evaluation of the results by independent assessors. In the most recent experiment (CASP14), deep-learning methods from one research group consistently delivered computed structures rivaling the corresponding experimental ones in accuracy. In this sense, the results represent a solution to the classical protein-folding problem, at least for single proteins. The models have already been shown to be capable of providing solutions for problematic crystal structures, and there are broad implications for the rest of structural biology. Other research groups also substantially improved performance. Here, we describe these results and outline some of the many implications. Other related areas of CASP, including modeling of protein complexes, structure refinement, estimation of model accuracy, and prediction of inter-residue contacts and distances, are also described.

Novel Reagent Space: Identifying Unorderable but Readily Synthesizable Building Blocks

Drug discovery building blocks available commercially or within an internal inventory cover a diverse range of chemical space and yet describe only a tiny fraction of all chemically feasible reagents. Vendors will eagerly provide tools to search the former; there is no straightforward method of mining the latter. We describe a procedure and use case in assembling chemical structures not available for purchase but that could likely be synthesized in one robust chemical transformation starting from readily available building blocks. Accessing this vast virtual chemical space dramatically increases our curated collection of reagents available for medicinal chemistry exploration and novel hit generation, almost tripling the number of those with 10 or fewer atoms.

Crystal structure prediction of materials with high symmetry using differential evolution

Crystal structure determines properties of materials. With the crystal structure of a chemical substance, many physical and chemical properties can be predicted by first-principles calculations or machine learning models. Since it is relatively easy to generate a hypothetical chemically valid formula, crystal structure prediction becomes an important method for discovering new materials. In our previous work, we proposed a contact map-based crystal structure prediction method, which uses global optimization algorithms such as genetic algorithms to maximize the match between the contact map of the predicted structure and the contact map of the real crystal structure to search for the coordinates at the Wyckoff positions (WP), demonstrating that known geometric constraints (such as the contact map of the crystal structure) help the crystal structure reconstruction. However, when predicting the crystal structure with high symmetry, we found that the global optimization algorithm has difficulty to find an effective combination of WP that satisfies the chemical formula, which is mainly caused by the inconsistency between the dimensionality of the contact map of the predicted crystal structure and the dimensionality of the contact map of the target crystal structure. This makes it challenging to predict the crystal structures of high-symmetry crystals. In order to solve this problem, here we propose to use PyXtal to generate and filter random crystal structures with given symmetry constraints based on the information such as chemical formulas and space groups. With contact map as the optimization goal, we use differential evolution algorithms to search for non-special coordinates at the WP to realize the structure prediction of high-symmetry crystal materials. Our experimental results show that our proposed algorithm CMCrystalHS can effectively solve the problem of inconsistent contact map dimensions and predict the crystal structures with high symmetry.

Machine Learning in Chemical Product Engineering: The State of the Art and a Guide for Newcomers

Chemical Product Engineering (CPE) is marked by numerous challenges, such as the complexity of the properties–structure–ingredients–process relationship of the different products and the necessity to discover and develop constantly and quickly new molecules and materials with tailor-made properties. In recent years, artificial intelligence (AI) and machine learning (ML) methods have gained increasing attention due to their performance in tackling particularly complex problems in various areas, such as computer vision and natural language processing. As such, they present a specific interest in addressing the complex challenges of CPE. This article provides an updated review of the state of the art regarding the implementation of ML techniques in different types of CPE problems with a particular focus on four specific domains, namely the design and discovery of new molecules and materials, the modeling of processes, the prediction of chemical reactions/retrosynthesis and the support for sensorial analysis. This review is further completed by general guidelines for the selection of an appropriate ML technique given the characteristics of each problem and by a critical discussion of several key issues associated with the development of ML modeling approaches. Accordingly, this paper may serve both the experienced researcher in the field as well as the newcomer.

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

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

Expanding the Chemical Information Science gateway

As chemical information evolves, impacting many chemistry areas, effective ways to disseminate results by the scientific community are also changing. Thus, publication schemes adapt to meet the needs of researchers across disciplines to share high-quality data, information, and knowledge. Since 2015, the F1000Research Chemical Information Science (CIS) gateway has offered an open and unique model to disseminate science at the interface of chemoinformatics, bioinformatics, and several other informatic-related disciplines. In response to the evolution of chemical information science, the F1000Research CIS gateway has incorporated new members to the advisory board. It is also reinforcing and expanding the gateway areas with a particular focus on machine learning and metabolomics. The range of available article types, availability of data, exposure within complementary multidisciplinary F1000Research gateways, and indexing in major bibliographic databases increases the visibility of all contributions. As part of progressing open science in this field, we look forward to your high-quality contributions to the CIS gateway.

Mastering the Gram-negative bacterial barrier - Chemical approaches to increase bacterial bioavailability of antibiotics

To win the battle against resistant, pathogenic bacteria, novel classes of anti-infectives and targets are urgently needed. Bacterial uptake, distribution, metabolic and efflux pathways of antibiotics in Gram-negative bacteria determine what we here refer to as bacterial bioavailability. Understanding these mechanisms from a chemical perspective is essential for anti-infective activity and hence, drug discovery as well as drug delivery. A systematic and critical discussion of in bacterio, in vitro and in silico assays reveals that a sufficiently accurate holistic approach is still missing. We expect new findings based on Gram-negative bacterial bioavailability to guide future anti-infective research.

Prediction of Chromatography Conditions for Purification in Organic Synthesis Using Deep Learning

In this research, a process for developing normal-phase liquid chromatography solvent systems has been proposed. In contrast to the development of conditions via thin-layer chromatography (TLC), this process is based on the architecture of two hierarchically connected neural network-based components. Using a large database of reaction procedures allows those two components to perform an essential role in the machine-learning-based prediction of chromatographic purification conditions, i.e., solvents and the ratio between solvents. In our paper, we build two datasets and test various molecular vectorization approaches, such as extended-connectivity fingerprints, learned embedding, and auto-encoders along with different types of deep neural networks to demonstrate a novel method for modeling chromatographic solvent systems employing two neural networks in sequence. Afterward, we present our findings and provide insights on the most effective methods for solving prediction tasks. Our approach results in a system of two neural networks with long short-term memory (LSTM)-based auto-encoders, where the first predicts solvent labels (by reaching the classification accuracy of 0.950 ± 0.001) and in the case of two solvents, the second one predicts the ratio between two solvents (R2 metric equal to 0.982 ± 0.001). Our approach can be used as a guidance instrument in laboratories to accelerate scouting for suitable chromatography conditions.

Quantum-mechanical property prediction of solvated drug molecules: what have we learned from a decade of SAMPL blind prediction challenges?

Joint academic-industrial projects supporting drug discovery are frequently pursued to deploy and benchmark cutting-edge methodical developments from academia in a real-world industrial environment at different scales. The dimensionality of tasks ranges from small molecule physicochemical property assessment over protein-ligand interaction up to statistical analyses of biological data. This way, method development and usability both benefit from insights gained at both ends, when predictiveness and readiness of novel approaches are confirmed, but the pharmaceutical drug makers get early access to novel tools for the quality of drug products and benefit of patients. Quantum-mechanical and simulation methods particularly fall into this group of methods, as they require skills and expense in their development but also significant resources in their application, thus are comparatively slowly dripping into the realm of industrial use. Nevertheless, these physics-based methods are becoming more and more useful. Starting with a general overview of these and in particular quantum-mechanical methods for drug discovery we review a decade-long and ongoing collaboration between Sanofi and the Kast group focused on the application of the embedded cluster reference interaction site model (EC-RISM), a solvation model for quantum chemistry, to study small molecule chemistry in the context of joint participation in several SAMPL (Statistical Assessment of Modeling of Proteins and Ligands) blind prediction challenges. Starting with early application to tautomer equilibria in water (SAMPL2) the methodology was further developed to allow for challenge contributions related to predictions of distribution coefficients (SAMPL5) and acidity constants (SAMPL6) over the years. Particular emphasis is put on a frequently overlooked aspect of measuring the quality of models, namely the retrospective analysis of earlier datasets and predictions in light of more recent and advanced developments. We therefore demonstrate the performance of the current methodical state of the art as developed and optimized for the SAMPL6 pKa and octanol-water log P challenges when re-applied to the earlier SAMPL5 cyclohexane-water log D and SAMPL2 tautomer equilibria datasets. Systematic improvement is not consistently found throughout despite the similarity of the problem class, i.e. protonation reactions and phase distribution. Hence, it is possible to learn about hidden bias in model assessment, as results derived from more elaborate methods do not necessarily improve quantitative agreement. This indicates the role of chance or coincidence for model development on the one hand which allows for the identification of systematic error and opportunities toward improvement and reveals possible sources of experimental uncertainty on the other. These insights are particularly useful for further academia-industry collaborations, as both partners are then enabled to optimize both the computational and experimental settings for data generation.

Machine learning-accelerated quantum mechanics-based atomistic simulations for industrial applications

Atomistic simulations have become an invaluable tool for industrial applications ranging from the optimization of protein-ligand interactions for drug discovery to the design of new materials for energy applications. Here we review recent advances in the use of machine learning (ML) methods for accelerated simulations based on a quantum mechanical (QM) description of the system. We show how recent progress in ML methods has dramatically extended the applicability range of conventional QM-based simulations, allowing to calculate industrially relevant properties with enhanced accuracy, at reduced computational cost, and for length and time scales that would have otherwise not been accessible. We illustrate the benefits of ML-accelerated atomistic simulations for industrial R&D processes by showcasing relevant applications from two very different areas, drug discovery (pharmaceuticals) and energy materials. Writing from the perspective of both a molecular and a materials modeling scientist, this review aims to provide a unified picture of the impact of ML-accelerated atomistic simulations on the pharmaceutical, chemical, and materials industries and gives an outlook on the exciting opportunities that could emerge in the future.

The challenges of generalizability in artificial intelligence for ADME/Tox endpoint and activity prediction

INTRODUCTION

Artificial intelligence (AI) has seen a massive resurgence in recent years with wide successes in computer vision, natural language processing, and games. The similar creation of robust and accurate AI models for ADME/Tox endpoint and activity prediction would be revolutionary to drug discovery pipelines. There have been numerous demonstrations of successful applications, but a key challenge remains: how generalizable are these predictive models?

AREAS COVERED

The authors present a summary of current promising components of AI models in the context of early drug discovery where ADME/Tox endpoint and activity prediction is the main driver of the iterative drug design process. Following that is a review of applicability domains and dataset construction considerations which determine generalizability bottlenecks for AI deployment. Further reviewed is the role of promising learning frameworks - multitask, transfer, and meta learning - which leverage auxiliary data to overcome issues of generalizability.

EXPERT OPINION

The authors conclude that the most promising direction toward integrating reliable and informative AI models into the drug discovery pipeline is a conjunction of learned feature representations, deep learning, and novel learning frameworks. Such a solution would address the sparse and incomplete datasets that are available for key endpoints related to drug discovery.

Chemical-informatics approach to COVID-19 drug discovery: Exploration of important fragments and data mining based prediction of some hits from natural origins as main protease (Mpro) inhibitors

As the world struggles against current global pandemic of novel coronavirus disease (COVID-19), it is challenging to trigger drug discovery efforts to search broad-spectrum antiviral agents. Thus, there is a need of strong and sustainable global collaborative works especially in terms of new and existing data analysis and sharing which will join the dots of knowledge gap. Our present chemical-informatics based data analysis approach is an attempt of application of previous activity data of SARS-CoV main protease (Mpro) inhibitors to accelerate the search of present SARS-CoV-2 Mpro inhibitors. The study design was composed of three major aspects: (1) classification QSAR based data mining of diverse SARS-CoV Mpro inhibitors, (2) identification of favourable and/or unfavourable molecular features/fingerprints/substructures regulating the Mpro inhibitory properties, (3) data mining based prediction to validate recently reported virtual hits from natural origin against SARS-CoV-2 Mpro enzyme. Our Structural and physico-chemical interpretation (SPCI) analysis suggested that heterocyclic nucleus like diazole, furan and pyridine have clear positive contribution while, thiophen, thiazole and pyrimidine may exhibit negative contribution to the SARS-CoV Mpro inhibition. Several Monte Carlo optimization based QSAR models were developed and the best model was used for screening of some natural product hits from recent publications. The resulted active molecules were analysed further from the aspects of fragment analysis. This approach set a stage for fragment exploration and QSAR based screening of active molecules against putative SARS-CoV-2 Mpro enzyme. We believe the future in vitro and in vivo studies would provide more perspectives for anti-SARS-CoV-2 agents.

Application of Artificial Intelligence in Medicine: An Overview

Artificial intelligence (AI) is a new technical discipline that uses computer technology to research and develop the theory, method, technique, and application system for the simulation, extension, and expansion of human intelligence. With the assistance of new AI technology, the traditional medical environment has changed a lot. For example, a patient's diagnosis based on radiological, pathological, endoscopic, ultrasonographic, and biochemical examinations has been effectively promoted with a higher accuracy and a lower human workload. The medical treatments during the perioperative period, including the preoperative preparation, surgical period, and postoperative recovery period, have been significantly enhanced with better surgical effects. In addition, AI technology has also played a crucial role in medical drug production, medical management, and medical education, taking them into a new direction. The purpose of this review is to introduce the application of AI in medicine and to provide an outlook of future trends.

Transmol: repurposing a language model for molecular generation

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

A novel molecular generation pipeline employing an attention-based neural network.

Artificial intelligence in drug discovery: what is realistic, what are illusions? Part 1: Ways to make an impact, and why we are not there yet

Although artificial intelligence (AI) has had a profound impact on areas such as image recognition, comparable advances in drug discovery are rare. This article quantifies the stages of drug discovery in which improvements in the time taken, success rate or affordability will have the most profound overall impact on bringing new drugs to market. Changes in clinical success rates will have the most profound impact on improving success in drug discovery; in other words, the quality of decisions regarding which compound to take forward (and how to conduct clinical trials) are more important than speed or cost. Although current advances in AI focus on how to make a given compound, the question of which compound to make, using clinical efficacy and safety-related end points, has received significantly less attention. As a consequence, current proxy measures and available data cannot fully utilize the potential of AI in drug discovery, in particular when it comes to drug efficacy and safety in vivo. Thus, addressing the questions of which data to generate and which end points to model will be key to improving clinically relevant decision-making in the future.

Activity Prediction of Small Molecule Inhibitors for Antirheumatoid Arthritis Targets Based on Artificial Intelligence

Rheumatoid arthritis (RA) is a chronic autoimmune disease, which is compared to "immortal cancer" in industry. Currently, SYK, BTK, and JAK are the three major targets of protein tyrosine kinase for this disease. According to existing research, marketed and research drugs for RA are mostly based on single target, which limits their efficacy. Therefore, designing multitarget or dual-target inhibitors provide new insights for the treatment of RA regarding of the specific association between SYK, BTK, and JAK from two signal transduction pathways. In this study, machine learning (XGBoost, SVM) and deep learning (DNN) models were combined for the first time to build a powerful integrated model for SYK, BTK, and JAK. The predictive power of the integrated model was proved to be superior to that of a single classifier. In order to accurately assess the generalization ability of the integrated model, comprehensive similarity analysis was performed on the training and the test set, and the prediction accuracy of the integrated model was specifically analyzed under different similarity thresholds. External validation was conducted using single-target and dual-target inhibitors, respectively. Results showed that our model not only obtained a high recall rate (97%) in single-target prediction, but also achieved a favorable yield (54.4%) in dual-target prediction. Furthermore, by clustering dual-target inhibitors, the prediction performance of model in various classes were proved, evaluating the applicability domain of the model in the dual-target drug screening. In summary, the integrated model proposed is promising to screen dual-target inhibitors of SYK/JAK or BTK/JAK as RA drugs, which is beneficial for the clinical treatment of rheumatoid arthritis.

Generating Multibillion Chemical Space of Readily Accessible Screening Compounds

An approach to the generation of ultra-large chemical libraries of readily accessible (“REAL”) compounds is described. The strategy is based on the use of two- or three-step three-component reaction sequences and available starting materials with pre-validated chemical reactivity. After the preliminary parallel experiments, the methods with at least ∼80% synthesis success rate (such as acylation – deprotection – acylation of monoprotected diamines or amide formation – click reaction with functionalized azides) can be selected and used to generate the target chemical space. It is shown that by using only on the two aforementioned reaction sequences, a nearly 29-billion compound library is easily obtained. According to the predicted physico-chemical descriptor values, the generated chemical space contains large fractions of both drug-like and “beyond rule-of-five” members, whereas the strictest lead-likeness criteria (the so-called Churcher's rules) are met by the lesser part, which still exceeds 22 million.

•

A strategy for ultra-large readily accessible (REAL) compound libraries is described

•

Pre-validated two- or three-step three-component reaction sequences are used

•

A 29-billion chemical space with ∼80% synthesis success rate has been easily obtained

Chemistry in Times of Artificial Intelligence

Chemists have to a large extent gained their knowledge by doing experiments and thus gather data. By putting various data together and then analyzing them, chemists have fostered their understanding of chemistry. Since the 1960s, computer methods have been developed to perform this process from data to information to knowledge. Simultaneously, methods were developed for assisting chemists in solving their fundamental questions such as the prediction of chemical, physical, or biological properties, the design of organic syntheses, and the elucidation of the structure of molecules. This eventually led to a discipline of its own: chemoinformatics. Chemoinformatics has found important applications in the fields of drug discovery, analytical chemistry, organic chemistry, agrichemical research, food science, regulatory science, material science, and process control. From its inception, chemoinformatics has utilized methods from artificial intelligence, an approach that has recently gained more momentum.

Practical considerations for active machine learning in drug discovery

Active machine learning enables the automated selection of the most valuable next experiments to improve predictive modelling and hasten active retrieval in drug discovery. Although a long established theoretical concept and introduced to drug discovery approximately 15 years ago, the deployment of active learning technology in the discovery pipelines across academia and industry remains slow. With the recent re-discovered enthusiasm for artificial intelligence as well as improved flexibility of laboratory automation, active learning is expected to surge and become a key technology for molecular optimizations. This review recapitulates key findings from previous active learning studies to highlight the challenges and opportunities of applying adaptive machine learning to drug discovery. Specifically, considerations regarding implementation, infrastructural integration, and expected benefits are discussed. By focusing on these practical aspects of active learning, this review aims at providing insights for scientists planning to implement active learning workflows in their discovery pipelines.

Models of necessity

The way chemists represent chemical structures as two-dimensional sketches made up of atoms and bonds, simplifying the complex three-dimensional molecules comprising nuclei and electrons of the quantum mechanical description, is the everyday language of chemistry. This language uses models, particularly of bonding, that are not contained in the quantum mechanical description of chemical systems, but has been used to derive machine-readable formats for storing and manipulating chemical structures in digital computers. This language is fuzzy and varies from chemist to chemist but has been astonishingly successful and perhaps contributes with its fuzziness to the success of chemistry. It is this creative imagination of chemical structures that has been fundamental to the cognition of chemistry and has allowed thought experiments to take place. Within the everyday language, the model nature of these concepts is not always clear to practicing chemists, so that controversial discussions about the merits of alternative models often arise. However, the extensive use of artificial intelligence (AI) and machine learning (ML) in chemistry, with the aim of being able to make reliable predictions, will require that these models be extended to cover all relevant properties and characteristics of chemical systems. This, in turn, imposes conditions such as completeness, compactness, computational efficiency and non-redundancy on the extensions to the almost universal Lewis and VSEPR bonding models. Thus, AI and ML are likely to be important in rationalizing, extending and standardizing chemical bonding models. This will not affect the everyday language of chemistry but may help to understand the unique basis of chemical language.

cheML.io: an online database of ML-generated molecules

Several recent ML algorithms for de novo molecule generation have been utilized to create an open-access database of virtual molecules. The algorithms were trained on samples from ZINC, a free database of commercially available compounds. Generated molecules, stemming from 10 different ML frameworks, along with their calculated properties were merged into a database and coupled to a web interface, which allows users to browse the data in a user friendly and convenient manner. ML-generated molecules with desired structures and properties can be retrieved with the help of a drawing widget. For the case of a specific search leading to insufficient results, users are able to create new molecules on demand. These newly created molecules will be added to the existing database and as a result, the content as well as the diversity of the database keeps growing in line with the user's requirements.

Several recent ML algorithms for de novo molecule generation have been utilized to create an open-access database of virtual molecules.