Efficient cocrystal coformer screening based on a Machine learning Strategy: A case study for the preparation of imatinib cocrystal with enhanced physicochemical properties

Cocrystal engineering, which involves the self-assembly of two or more components into a solid-state supramolecular structure through non-covalent interactions, has emerged as a promising approach to tailor the physicochemical properties of active pharmaceutical ingredient (API). Efficient coformer screening for cocrystal remains a challenge. Herein, a prediction strategy based on machine learning algorithms was employed to predict cocrystal formation and seven reliable models with accuracy over 0.890 were successfully constructed. Imatinib was selected as the model drug and the models established were applied to screen 31 potential coformers. Experimental verification results indicated RF-8 is the optimal model among seven models with an accuracy of 0.839. When the seven models were combined for coformer screening of Imatinib, the combinational model achieved an accuracy of 0.903, and eight new solid forms were observed and characterized. Benefiting from intermolecular interactions, the obtained multicomponent crystals displayed enhanced physicochemical properties. Dissolution and solubility experiments showed the prepared multicomponent crystals had higher cumulative dissolution rate and remarkably improved the solubility of imatinib, and IM-MC exhibited comparable solubility to Imatinib mesylate α form. Stability test and cytotoxicity results showed that multicomponent crystals exhibited excellent stability and the drug-drug cocrystal IM-5F exhibited higher cytotoxicity than pure API.

Thermodynamic Stability Is a Poor Indicator of Cocrystallization in Models of Organic Molecules

Cocrystallizing a given molecule with another can be useful for adjusting the physical properties of molecules in the solid state. However, most combinations of molecules do not readily cocrystallize but form either one-component crystals or amorphous solids. Computational methods of crystal structure prediction can, in principle, identify the thermodynamically stable cocrystal and thus predict if molecules will cocrystallize or not. However, the pronounced polymorphism and tendency of many organic molecules to form disordered solids suggest that kinetic factors can play an important role in cocrystallization. The question remains: if a binary system of molecules has a thermodynamically stable cocrystal, will it indeed cocrystallize? To address this question, we simulate the crystallization of more than 2600 distinct pairs of chiral model molecules of similar size in 2D and calculate accurate crystal energy landscapes for all of them. Our analysis shows that thermodynamic criteria alone are unreliable in the prediction of cocrystallization. While the vast majority of cocrystals that form in our simulations are thermodynamically favorable, most coformer systems that have a thermodynamically stable cocrystal do not cocrystallize. We furthermore show that cocrystallization rates increase 3-fold when coformers are used that do not form well-ordered single-component crystals. Our results suggest that kinetic factors of cocrystallization are decisive in many cases.

In silico co-crystal design: Assessment of the latest advances

Pharmaceutical co-crystals represent a growing class of crystal forms in the context of pharmaceutical science. They are attractive to pharmaceutical scientists because they significantly expand the number of crystal forms that exist for an active pharmaceutical ingredient and can lead to improvements in physicochemical properties of clinical relevance. At the same time, machine learning is finding its way into all areas of drug discovery and delivers impressive results. In this review, we attempt to provide an overview of machine learning, deep learning and network-based recommendation approaches applied to pharmaceutical co-crystallization. We also present crystal structure prediction as an alternative to machine learning approaches.

Metronidazole Cocrystal Polymorphs with Gallic and Gentisic Acid Accessed through Slurry, Atomization Techniques, and Thermal Methods

In this study, key features of metronidazole (MNZ) cocrystal polymorphs with gallic acid (GAL) and gentisic acid (GNT) were elucidated. Solvent-mediated phase transformation experiments in 30 solvents with varying properties were employed to control the polymorphic behavior of the MNZ cocrystal with GAL. Solvents with relative polarity (RP) values above 0.35 led to cocrystal I°, the thermodynamically stable form. Conversely, solvents with RP values below 0.35 produced cocrystal II, which was found to be only 0.3 kJ mol-1 less stable in enthalpy. The feasibility of electrospraying, including solvent properties and process conditions required, and spray drying techniques to control cocrystal polymorphism was also investigated, and these techniques were found to facilitate exclusive formation of the metastable MNZ-GAL cocrystal II. Additionally, the screening approach resulted in a new, high-temperature polymorph I of the MNZ-GNT cocrystal system, which is enantiotropically related to the already known form II°. The intermolecular energy calculations, as well as the 2D similarity between the MNZ-GAL polymorphs and the 3D similarity between MNZ-GNT polymorphs, rationalized the observed transition behaviors. Furthermore, the evaluation of virtual cocrystal screening techniques identified molecular electrostatic potential calculations as a supportive tool for coformer selection.

Recent Advances in Co-Former Screening and Formation Prediction of Multicomponent Solid Forms of Low Molecular Weight Drugs

Multicomponent solid forms of low molecular weight drugs, such as co-crystals, salts, and co-amorphous systems, are a result of the combination of an active pharmaceutical ingredient (API) with a pharmaceutically acceptable co-former. These solid forms can enhance the physicochemical and pharmacokinetic properties of APIs, making them increasingly interesting and important in recent decades. Nevertheless, predicting the formation of API multicomponent solid forms in the early stages of formulation development can be challenging, as it often requires significant time and resources. To address this, empirical and computational methods have been developed to help screen for potential co-formers more efficiently and accurately, thus reducing the number of laboratory experiments needed. This review provides a comprehensive overview of current screening and prediction methods for the formation of API multicomponent solid forms, covering both crystalline states (co-crystals and salts) and amorphous forms (co-amorphous). Furthermore, it discusses recent advances and emerging trends in prediction methods, with a particular focus on artificial intelligence.

Cocrystal Synthesis through Crystal Structure Prediction

Crystal structure prediction (CSP) is an invaluable tool in the pharmaceutical industry because it allows to predict all the possible crystalline solid forms of small-molecule active pharmaceutical ingredients. We have used a CSP-based cocrystal prediction method to rank ten potential cocrystal coformers by the energy of the cocrystallization reaction with an antiviral drug candidate, MK-8876, and a triol process intermediate, 2-ethynylglyclerol. For MK-8876, the CSP-based cocrystal prediction was performed retrospectively and successfully predicted the maleic acid cocrystal as the most likely cocrystal to be observed. The triol is known to form two different cocrystals with 1,4-diazabicyclo[2.2.2]octane (DABCO), but a larger solid form landscape was desired. CSP-based cocrystal screening predicted the triol-DABCO cocrystal as rank one, while a triol-l-proline cocrystal was predicted as rank two. Computational finite-temperature corrections enabled determination of relative crystallization propensities of the triol-DABCO cocrystals with different stoichiometries and prediction of the triol-l-proline polymorphs in the free-energy landscape. The triol-l-proline cocrystal was obtained during subsequent targeted cocrystallization experiments and was found to exhibit an improved melting point and deliquescence behavior over the triol-free acid, which could be considered as an alternative solid form in the synthesis of islatravir.

Absolute Configuration Determination of Chiral API Molecules by MicroED Analysis of Cocrystal Powders Formed Based on Cocrystal Propensity Prediction Calculations

Establishing the absolute configuration of chiral active pharmaceutical ingredients (APIs) is of great importance. Single crystal X-ray diffraction (scXRD) has traditionally been the method of choice for such analysis, but scXRD requires the growth of large crystals, which can be challenging. Here, we present a method for determining absolute configuration that does not rely on the growth of large crystals. By examining microcrystals formed with chiral probes (small chiral compounds such as amino acids), absolute configuration can be unambiguously determined by microcrystal electron diffraction (MicroED). Our streamlined method employs three steps: (1) virtual screening to identify promising chiral probes, (2) experimental cocrystal screening and (3) structure determination by MicroED and absolute configuration assignment. We successfully applied this method to analyze two chiral API molecules currently on the market for which scXRD was not used to determine absolute configuration.

Virtual Screening, Structural Analysis, and Formation Thermodynamics of Carbamazepine Cocrystals

In this study, the existing set of carbamazepine (CBZ) cocrystals was extended through the successful combination of the drug with the positional isomers of acetamidobenzoic acid. The structural and energetic features of the CBZ cocrystals with 3- and 4-acetamidobenzoic acids were elucidated via single-crystal X-ray diffraction followed by QTAIMC analysis. The ability of three fundamentally different virtual screening methods to predict the correct cocrystallization outcome for CBZ was assessed based on the new experimental results obtained in this study and data available in the literature. It was found that the hydrogen bond propensity model performed the worst in distinguishing positive and negative results of CBZ cocrystallization experiments with 87 coformers, attaining an accuracy value lower than random guessing. The method that utilizes molecular electrostatic potential maps and the machine learning approach named CCGNet exhibited comparable results in terms of prediction metrics, albeit the latter resulted in superior specificity and overall accuracy while requiring no time-consuming DFT computations. In addition, formation thermodynamic parameters for the newly obtained CBZ cocrystals with 3- and 4-acetamidobenzoic acids were evaluated using temperature dependences of the cocrystallization Gibbs energy. The cocrystallization reactions between CBZ and the selected coformers were found to be enthalpy-driven, with entropy terms being statistically different from zero. The observed difference in dissolution behavior of the cocrystals in aqueous media was thought to be caused by variations in their thermodynamic stability.

Exploring the Cocrystal Landscape of Posaconazole by Combining High-Throughput Screening Experimentation with Computational Chemistry

The development of multicomponent crystal forms, such as cocrystals, represents a means to enhance the dissolution and absorption properties of poorly water-soluble drug compounds. However, the successful discovery of new pharmaceutical cocrystals remains a time- and resource-consuming process. This study proposes the use of a combined computational-experimental high-throughput approach as a tool to accelerate and improve the efficiency of cocrystal screening exemplified by posaconazole. First, we employed the COSMOquick software to preselect and rank cocrystal candidates (coformers). Second, high-throughput crystallization experiments (HTCS) were conducted on the selected coformers. The HTCS results were successfully reproduced by liquid-assisted grinding and reaction crystallization, ultimately leading to the synthesis of thirteen new posaconazole cocrystals (7 anhydrous, 5 hydrates, and 1 solvate). The posaconazole cocrystals were characterized by PXRD, ¹H NMR, Fourier transform-Raman, thermogravimetry-Fourier transform infrared spectroscopy, and differential scanning calorimetry. In addition, the prediction performance of COSMOquick was compared to that of two alternative knowledge-based methods: molecular complementarity (MC) and hydrogen bond propensity (HBP). Although HBP does not perform better than random guessing for this case study, both MC and COSMOquick show good discriminatory ability, suggesting their use as a potential virtual tool to improve cocrystal screening.

Gefitinib-resveratrol Cocrystal with Optimized Performance in Dissolution and Stability

Gefitinib (GEF) is an anti-tumor oral solid formulation with a superior advantage for lung tumors. However, it has poor aqueous solubility which limits its utility in vivo. Herein, a novel cocrystal (GEF-RES) assembled by GEF and RES (Resveratrol) has been successfully prepared and comprehensively characterized by differential scanning calorimetry, thermogravimetric analysis, Raman spectroscopy and powder X-ray diffraction. A single-crystal structure of the GEF-RES cocrystal was solved and illustrated in detail. In aqueous hydrochloric acid, the GEF-RES cocrystal showed that the maximum concentration of GEF was slightly higher than that of raw GEF. Furthermore, the thermal and physical stability of the GEF-RES cocrystal were also evaluated in this paper. The enhanced solubility and excellent solid-state stability results may provide new potential to the application of key GEF in clinical.

The First Noncovalent-Bonded Supramolecular Frameworks of (Benzylthio)Acetic Acid with Proline Compounds, Isonicotinamide and Tryptamine

The co-crystallization of (benzylthio)acetic acid (HBTA) with L-proline (L-PRO), D-proline (D-PRO), DL-proline (DL-PRO), isonicotinamide (INA) and tryptamine (TPA) led to the formation of five novel crystalline compounds: L-PRO^±·HBTA (1), D-PRO^±·HBTA (2), DL-PRO^±·HBTA (3), INA·HBTA (4) and TPA⁺·BTA^- (5). The prepared supramolecular assemblies were characterized by single crystal X-ray diffraction, an elemental analysis, FT-IR spectroscopy and a thermal analysis based on thermogravimetry (TG) combined with differential scanning calorimetry (DSC). Additionally, their melting points through TG/DSC measurements were established. All fabricated adducts demonstrated the same stoichiometry, displayed as 1:1. The integration of HBTA with selected N-containing co-formers yielded different forms of multi-component crystalline phases: zwitterionic co-crystals (1-3), true co-crystal (4) or true salt (5). In the asymmetric units of 1-4, the acidic ingredient is protonated, whereas the corresponding N-containing entities take either the zwitterionic form (1-3) or remain in the original neutral figure (4). The molecular structure of complex 5 is occupied by the real ionic forms of both components, namely the (benzylthio)acetate anion (BTA^-) and the tryptaminium cation (TPA⁺). In crystals 1-5, the respective molecular residues are permanently bound to each other via strong H-bonds provided by the following pairs of donor···acceptor: O_carboxylic···O_carboxylate and N_{pyrrolidinium}···O_carboxylate in 1-3, O_carboxylic···N_pyridine and N_amine···O_carboxylic in 4 as well as N_indole···O_carboxylate and N_aminium···O_carboxylate in 5. The crystal structures of conglomerates 1-5 are also stabilized by numerous weaker intermolecular contacts, including C-H···O (1-3, 5), C-H···S (1, 2, 5), C-H···N (5), C-H···C (5), C-H···π (1-5) as well as π···π (4) interactions. The different courses of registered FT-IR spectral traces and thermal profiles for materials 1-5 in relation to their counterparts, gained for the pure molecular ingredients, also clearly confirm the formation of new crystalline phases.

Cocrystal Prediction of Bexarotene by Graph Convolution Network and Bioavailability Improvement

Bexarotene (BEX) was approved by the FDA in 1999 for the treatment of cutaneous T-cell lymphoma (CTCL). The poor aqueous solubility causes the low bioavailability of the drug and thereby limits the clinical application. In this study, we developed a GCN-based deep learning model (CocrystalGCN) for in-silico screening of the cocrystals of BEX. The results show that our model obtained high performance relative to baseline models. The top 30 of 109 coformer candidates were scored by CocrystalGCN and then validated experimentally. Finally, cocrystals of BEX-pyrazine, BEX-2,5-dimethylpyrazine, BEX-methyl isonicotinate, and BEX-ethyl isonicotinate were successfully obtained. The crystal structures were determined by single-crystal X-ray diffraction. Powder X-ray diffraction, differential scanning calorimetry, and thermogravimetric analysis were utilized to characterize these multi-component forms. All cocrystals present superior solubility and dissolution over the parent drug. The pharmacokinetic studies show that the plasma exposures (AUC_0−8h) of BEX-pyrazine and BEX-2,5-dimethylpyrazine are 1.7 and 1.8 times that of the commercially available BEX powder, respectively. This work sets a good example for integrating virtual prediction and experimental screening to discover the new cocrystals of water-insoluble drugs.

Applications of Artificial Intelligence and Machine Learning Algorithms to Crystallization

Artificial intelligence and specifically machine learning applications are nowadays used in a variety of scientific applications and cutting-edge technologies, where they have a transformative impact. Such an assembly of statistical and linear algebra methods making use of large data sets is becoming more and more integrated into chemistry and crystallization research workflows. This review aims to present, for the first time, a holistic overview of machine learning and cheminformatics applications as a novel, powerful means to accelerate the discovery of new crystal structures, predict key properties of organic crystalline materials, simulate, understand, and control the dynamics of complex crystallization process systems, as well as contribute to high throughput automation of chemical process development involving crystalline materials. We critically review the advances in these new, rapidly emerging research areas, raising awareness in issues such as the bridging of machine learning models with first-principles mechanistic models, data set size, structure, and quality, as well as the selection of appropriate descriptors. At the same time, we propose future research at the interface of applied mathematics, chemistry, and crystallography. Overall, this review aims to increase the adoption of such methods and tools by chemists and scientists across industry and academia.

Emerging Landscape of Computational Modeling in Pharmaceutical Development

Computational chemistry applications have become an integral part of the drug discovery workflow over the past 35 years. However, computational modeling in support of drug development has remained a relatively uncharted territory for a significant part of both academic and industrial communities. This review considers the computational modeling workflows for three key components of drug preclinical and clinical development, namely, process chemistry, analytical research and development, as well as drug product and formulation development. An overview of the computational support for each step of the respective workflows is presented. Additionally, in context of solid form design, special consideration is given to modern physics-based virtual screening methods. This covers rational approaches to polymorph, coformer, counterion, and solvent virtual screening in support of solid form selection and design.

Novel Cocrystals of Vonoprazan: Machine Learning-Assisted Discovery

Vonoprazan (VPZ) is the first-in-class potassium-competitive acid blocker (P-CAB), and has many advantages over proton pump inhibitors (PPIs). It is administered as a fumarate salt for the treatment of acid-related diseases, including reflux esophagitis, gastric ulcer, and duodenal ulcer, and for eradication of Helicobacter pylori. To discover novel cocrystals of VPZ, we adopted an artificial neural network (ANN)-based machine learning model as a virtual screening tool that can guide selection of the most promising coformers for VPZ cocrystals. Experimental screening by liquid-assisted grinding (LAG) confirmed that 8 of 19 coformers selected by the ANN model were likely to create new solid forms with VPZ. Structurally similar benzenediols and benzenetriols, i.e., catechol (CAT), resorcinol (RES), hydroquinone (HYQ), and pyrogallol (GAL), were used as coformers to obtain phase pure cocrystals with VPZ by reaction crystallization. We successfully prepared and characterized three novel cocrystals: VPZ–RES, VPZ–CAT, and VPZ–GAL. VPZ–RES had the highest solubility among the novel cocrystals studied here, and was even more soluble than the commercially available fumarate salt of VPZ in solution at pH 6.8. In addition, novel VPZ cocrystals had superior stability in aqueous media than VPZ fumarates, demonstrating their potential for improved pharmaceutical performance.

A study to discover novel pharmaceutical cocrystals of pelubiprofen with a machine learning approach compared

Pelubiprofen (PF), a biopharmaceutical classification system (BCS) class II non-steroidal anti-inflammatory drug, has been on the market only in its crystalline form. To discover the first cocrystal form(s) of the...

Salts of 2-hydroxybenzylamine with improvements on solubility and stability: Virtual and experimental screening

2-Hydroxybenzylamine (2-HOBA) is a drug used to effectively treat oxidative stress. To improve its aqueous solubility and thermal stability, salt screening and synthesis was carried out. The conductor-like screening model for the real solvents model (COSMO-RS) was applied to virtual screening of coformers among 200 commonly used candidates for salification of 2-HOBA. As a result, 40 hit compounds were subjected to experimental liquid-assisted grinding (LAG) with 2-HOBA, then 21 systems were characterized as new solid phases by PXRD. Nine multicomponent single crystals of 2-HOBA with succinic acid, p-aminobenzoic acid, p-nitrobenzoic acid, o-nitrobenzoic acid, p-toluic acid, 2,3-dihydroxybenzoic acid, 3,4-dihydroxybenzoic acid, p-nitrophenol, and 5-hydroxyisophthalic acid were obtained and characterized by single-crystal X-ray diffraction, powder X-ray diffraction, differential scanning calorimetry, and thermogravimetric analysis. All of them were salts and exhibited higher decomposition temperatures compared with pure 2-HOBA. The apparent aqueous solubility of three new salts, i.e., those with succinic acid, p-aminobenzoic acid, and p-nitrophenol were higher than the equilibrium solubility of 2-HOBA. The accelerated stability test indicated that all salts show excellent stability under conditions (40 °C and 75% RH) for 4 weeks. Overall, this work introduced a protocol that combined the virtual screening tool based on the COSMO-RS model and the experimental LAG method to screen new salts for a target compound. The feasibility of this protocol was confirmed in the case of 2-HOBA whose new salts were successfully obtained and represented an improvement for aqueous solubility and thermal stability.

Coupling complementary strategy to flexible graph neural network for quick discovery of coformer in diverse co-crystal materials

Cocrystal engineering have been widely applied in pharmaceutical, chemistry and material fields. However, how to effectively choose coformer has been a challenging task on experiments. Here we develop a graph neural network (GNN) based deep learning framework to quickly predict formation of the cocrystal. In order to capture main driving force to crystallization from 6819 positive and 1052 negative samples reported by experiments, a feasible GNN framework is explored to integrate important prior knowledge into end-to-end learning on the molecular graph. The model is strongly validated against seven competitive models and three challenging independent test sets involving pharmaceutical cocrystals, π-π cocrystals and energetic cocrystals, exhibiting superior performance with accuracy higher than 96%, confirming its robustness and generalization. Furthermore, one new energetic cocrystal predicted is successfully synthesized, showcasing high potential of the model in practice. All the data and source codes are available at https://github.com/Saoge123/ccgnet for aiding cocrystal community.

SMINBR: An Integrated Network and Chemoinformatics Tool Specialized for Prediction of Two-Component Crystal Formation

Two-component crystals such as pharmaceutical cocrystals and salts have been proven as an effective strategy to improve physicochemical and biopharmaceutical properties of drugs. It is not easy to select proper molecular combinations to form two-component crystals. The network-based models have been successfully utilized to guide cocrystal design. Yet, the traditional social network-derived methods based on molecular-interaction topology information cannot directly predict interaction partners for new chemical entities (NCEs) that have not been observed to form two-component crystals. Herein, we proposed an effective tool, namely substructure-molecular-interaction network-based recommendation (SMINBR), to prioritize potential interaction partners for NCEs. This in silico tool incorporates network and chemoinformatics methods to bridge the gap between NCEs and known molecular-interaction network. The high performance of 10-fold cross validation and external validation shows the high accuracy and good generalization capability of the model. As a case study, top 10 recommended coformers for apatinib were all experimentally confirmed and a new apatinib cocrystal with paradioxybenzene was obtained. The predictive capability of the model attributes to its accordance with complementary patterns driving the formation of intermolecular interactions. SMINBR could automatically recommend new interaction partners for a target molecule, and would be an effective tool to guide cocrystal design. A free web server for SMINBR is available at http://lmmd.ecust.edu.cn/sminbr/.

Pulmonary Drug Delivery of Antimicrobials and Anticancer Drugs Using Solid Dispersions

It is well established that currently available inhaled drug formulations are associated with extremely low lung deposition. Currently available technologies alleviate this low deposition problem via mixing the drug with inert larger particles, such as lactose monohydrate. Those inert particles are retained in the inhalation device or impacted in the throat and swallowed, allowing the smaller drug particles to continue their journey towards the lungs. While this seems like a practical approach, in some formulations, the ratio between the carrier to drug particles can be as much as 30 to 1. This limitation becomes more critical when treating lung conditions that inherently require large doses of the drug, such as antibiotics and antivirals that treat lung infections and anticancer drugs. The focus of this review article is to review the recent advancements in carrier free technologies that are based on coamorphous solid dispersions and cocrystals that can improve flow properties, and help with delivering larger doses of the drug to the lungs.

Progress in Research on Artificial Intelligence Applied to Polymorphism and Cocrystal Prediction

Artificial intelligence (AI) is a technology that builds an artificial system with certain intelligence and uses computer software and hardware to simulate intelligent human behavior. When combined with drug research and development, AI can considerably shorten this cycle, improve research efficiency, and minimize costs. The use of machine learning to discover novel materials and predict material properties has become a new research direction. On the basis of the current status of worldwide research on the combination of AI and crystal form and cocrystal, this mini-review analyzes and explores the application of AI in polymorphism prediction, crystal structure analysis, crystal property prediction, cocrystal former (CCF) screening, cocrystal composition prediction, and cocrystal formation prediction. This study provides insights into the future applications of AI in related fields.

Cocrystals Based on 4,4’-bipyridine: Influence of Crystal Packing on Melting Point

The reactions of piperonylic acid (HPip) and cinnamic acid (HCinn) with 4,4’-bipyridine (4,4’-bipy) have been assayed using the same synthetic methodology, yielding two binary cocrystals with different acid:4,4’-bipy molar ratios, (HPip)(4,4’-bipy) (1) and (HCinn)2(4,4’-bipy) (2). The melting point (m.p.) of these cocrystals have been measured and a remarkable difference (ΔT ≈ 78 °C) between them was observed. Moreover, the two cocrystals have been characterized by powder X-ray diffraction (PXRD), elemental analysis (EA), FTIR-ATR, 1H NMR spectroscopies, and single-crystal X-ray diffraction. The study of their structural packings via Hirshfeld surface analysis and energy frameworks revealed the important contribution of the π···π and C-H···π interactions to the formation of different structural packing motifs, this being the main reason for the difference of m.p. between them. Moreover, it has been observed that 1 and 2 presented the same packing motifs as the crystal structure of their corresponding carboxylic acids, but 1 and 2 showed lower m.p. than those of the carboxylic acids, which could be related to the lower strength of the acid-pyridine heterosynthons respect to the acid-acid homosynthons in the crystal structures.

Cocrystal Prediction Using Machine Learning Models and Descriptors

Cocrystals are of much interest in industrial application as well as academic research, and screening of suitable coformers for active pharmaceutical ingredients is the most crucial and challenging step in cocrystal development. Recently, machine learning techniques are attracting researchers in many fields including pharmaceutical research such as quantitative structure-activity/property relationship. In this paper, we develop machine learning models to predict cocrystal formation. We extract descriptor values from simplified molecular-input line-entry system (SMILES) of compounds and compare the machine learning models by experiments with our collected data of 1476 instances. As a result, we found that artificial neural network shows great potential as it has the best accuracy, sensitivity, and F1 score. We also found that the model achieved comparable performance with about half of the descriptors chosen by feature selection algorithms. We believe that this will contribute to faster and more accurate cocrystal development.

Multicomponent crystals of clotrimazole: a combined theoretical and experimental study

Compared with clotrimazole, some multicomponent crystals showed an improvement in solubility and dissolution rate.

Virtual coformer screening by a combined machine learning and physics-based approach

Cocrystals as a solid form technology for improving physicochemical properties have gained increasing popularity in the pharmaceutical, nutraceutical, and agrochemical industries.

One class classification as a practical approach for accelerating π-π co-crystal discovery

The implementation of machine learning models has brought major changes in the decision-making process for materials design. One matter of concern for the data-driven approaches is the lack of negative data from unsuccessful synthetic attempts, which might generate inherently imbalanced datasets. We propose the application of the one-class classification methodology as an effective tool for tackling these limitations on the materials design problems. This is a concept of learning based only on a well-defined class without counter examples. An extensive study on the different one-class classification algorithms is performed until the most appropriate workflow is identified for guiding the discovery of emerging materials belonging to a relatively small class, that being the weakly bound polyaromatic hydrocarbon co-crystals. The two-step approach presented in this study first trains the model using all the known molecular combinations that form this class of co-crystals extracted from the Cambridge Structural Database (1722 molecular combinations), followed by scoring possible yet unknown pairs from the ZINC15 database (21 736 possible molecular combinations). Focusing on the highest-ranking pairs predicted to have higher probability of forming co-crystals, materials discovery can be accelerated by reducing the vast molecular space and directing the synthetic efforts of chemists. Further on, using interpretability techniques a more detailed understanding of the molecular properties causing co-crystallization is sought after. The applicability of the current methodology is demonstrated with the discovery of two novel co-crystals, namely pyrene-6H-benzo[c]chromen-6-one (1) and pyrene-9,10-dicyanoanthracene (2).

Machine learning using one class classification on a database of existing co-crystals enables the identification of co-formers which are likely to form stable co-crystals, resulting in the synthesis of two co-crystals of polyaromatic hydrocarbons.

Co-crystal Prediction by Artificial Neural Networks*

A significant amount of attention has been given to the design and synthesis of co-crystals by both industry and academia because of its potential to change a molecule's physicochemical properties. Yet, difficulties arise when searching for adequate combinations of molecules (or coformers) to form co-crystals, hampering the efficient exploration of the target's solid-state landscape. This paper reports on the application of a data-driven co-crystal prediction method based on two types of artificial neural network models and co-crystal data present in the Cambridge Structural Database. The models accept pairs of coformers and predict whether a co-crystal is likely to form. By combining the output of multiple models of both types, our approach shows to have excellent performance on the proposed co-crystal training and validation sets, and has an estimated accuracy of 80 % for molecules for which previous co-crystallization data is unavailable.

Virtual Coformer Screening by Crystal Structure Predictions: Crucial Role of Crystallinity in Pharmaceutical Cocrystallization

One of the most popular strategies of the optimization of drug properties in the pharmaceutical industry appears to be a solid form changing into a cocrystalline form. A number of virtual screening approaches have been previously developed to allow a selection of the most promising cocrystal formers (coformers) for an experimental follow-up. A significant drawback of those methods is related to the lack of accounting for the crystallinity contribution to cocrystal formation. To address this issue, we propose in this study two virtual coformer screening approaches based on a modern cloud-computing crystal structure prediction (CSP) technology at a dispersion-corrected density functional theory (DFT-D) level. The CSP-based methods were for the first time validated on challenging cases of indomethacin and paracetamol cocrystallization, for which the previously developed approaches provided poor predictions. The calculations demonstrated a dramatic improvement of the virtual coformer screening performance relative to the other methods. It is demonstrated that the crystallinity contribution to the formation of paracetamol and indomethacin cocrystals is a dominant one and, therefore, should not be ignored in the virtual screening calculations. Our results encourage a broad utilization of the proposed CSP-based technology in the pharmaceutical industry as the only virtual coformer screening method that directly accounts for the crystallinity contribution.

A Short Review of Current Computational Concepts for High-Pressure Phase Transition Studies in Molecular Crystals

High-pressure chemistry of organic compounds is a hot topic of modern chemistry. In this work, basic computational concepts for high-pressure phase transition studies in molecular crystals are described, showing their advantages and disadvantages. The interconnection of experimental and computational methods is highlighted, showing the importance of energy calculations in this field. Based on our deep understanding of methods’ limitations, we suggested the most convenient scheme for the computational study of high-pressure crystal structure changes. Finally, challenges and possible ways for progress in high-pressure phase transitions research of organic compounds are briefly discussed.

Identification of a previously unreported co-crystal form of acetazolamide: a combination of multiple experimental and virtual screening methods

In this work, we demonstrate an approach of trying multiple methods in a more comprehensive search for co-crystals of acetazolamide.

Systematic coformer contribution to cocrystal stabilization: energy and packing trends

CSD data mining and energy calculations show that coformer self-interactions might significantly contribute to the packing energy stabilization of cocrystals.

Understanding the formation of apremilast cocrystals

Apremilast (APR), an anti-psoriatic agent, easily forms isostructural cocrystals and solvates with aromatic entities, often disobeying at the same time Kitaigorodsky's rule as to the saturation of possible hydrogen-bonding sites. In this paper the reasons for this peculiar behavior are investigated, employing a joint experimental and theoretical approach. This includes the design of cocrystals with coformers having a high propensity towards the formation of both aromatic-aromatic and hydrogen-bonding interactions, determination of their structure, using solid-state NMR spectroscopy and X-ray crystallography, as well as calculations of stabilization energies of formation of the obtained cocrystals, followed by crystal structure prediction calculations and solubility measurements. The findings indicate that the stabilization energies of cocrystal formation are positive in all cases, which results from strain in the APR conformation in these crystal forms. On the other hand, solubility measurements show that the Gibbs free energy of formation of the apremilast:picolinamide cocrystal is negative, suggesting that the formation of the studied cocrystals is entropy driven. This entropic stabilization is associated with the disorder observed in almost all known cocrystals and solvates of APR.

Intermolecular Interactions in Functional Crystalline Materials: From Data to Knowledge

Intermolecular interactions of organic, inorganic, and organometallic compounds are the key to many composition–structure and structure–property networks. In this review, some of these relations and the tools developed by the Cambridge Crystallographic Data Center (CCDC) to analyze them and design solid forms with desired properties are described. The potential of studies supported by the Cambridge Structural Database (CSD)-Materials tools for investigation of dynamic processes in crystals, for analysis of biologically active, high energy, optical, (electro)conductive, and other functional crystalline materials, and for the prediction of novel solid forms (polymorphs, co-crystals, solvates) are discussed. Besides, some unusual applications, the potential for further development and limitations of the CCDC software are reported.

Cocrystals in the Cambridge Structural Database: a network approach

To obtain a better understanding of which coformers to combine for the successful formation of a cocrystal, techniques from data mining and network science are used to analyze the data contained in the Cambridge Structural Database (CSD). A network of coformers is constructed based on cocrystal entries present in the CSD and its properties are analyzed. From this network, clusters of coformers with a similar tendency to form cocrystals are extracted. The popularity of the coformers in the CSD is unevenly distributed: a small group of coformers is responsible for most of the cocrystals, hence resulting in an inherently biased data set. The coformers in the network are found to behave primarily in a bipartite manner, demonstrating the importance of combining complementary coformers for successful cocrystallization. Based on our analysis, it is demonstrated that the CSD coformer network is a promising source of information for knowledge-based cocrystal prediction.

Cocrystal design by network-based link prediction

Cocrystals are predicted using a network of coformers extracted from the CSD.

Evaluating the Energetic Driving Force for Cocrystal Formation

We present a periodic density functional theory study of the stability of 350 organic cocrystals relative to their pure single-component structures, the largest study of cocrystals yet performed with high-level computational methods. Our calculations demonstrate that cocrystals are on average 8 kJ mol^-1 more stable than their constituent single-component structures and are very rarely (<5% of cases) less stable; cocrystallization is almost always a thermodynamically favorable process. We consider the variation in stability between different categories of systems-hydrogen-bonded, halogen-bonded, and weakly bound cocrystals-finding that, contrary to chemical intuition, the presence of hydrogen or halogen bond interactions is not necessarily a good predictor of stability. Finally, we investigate the correlation of the relative stability with simple chemical descriptors: changes in packing efficiency and hydrogen bond strength. We find some broad qualitative agreement with chemical intuition-more densely packed cocrystals with stronger hydrogen bonding tend to be more stable-but the relationship is weak, suggesting that such simple descriptors do not capture the complex balance of interactions driving cocrystallization. Our conclusions suggest that while cocrystallization is often a thermodynamically favorable process, it remains difficult to formulate general rules to guide synthesis, highlighting the continued importance of high-level computation in predicting and rationalizing such systems.

Co-crystallization of anti-inflammatory pharmaceutical contaminants and rare carboxylic acid–pyridine supramolecular synthon breakdown

Co-crystallization of the pharmaceutical contaminants mefenamic acid and naproxen is reported; one co-crystal exhibits a rare carboxylic acid–pyridine synthon breakdown.