Evaluation of log P, pKa, and log D predictions from the SAMPL7 blind challenge

Multistate Method to Efficiently Account for Tautomerism and Protonation in Alchemical Free-Energy Calculations

The majority of drug-like molecules contain at least one ionizable group, and many common drug scaffolds are subject to tautomeric equilibria. Thus, these compounds are found in a mixture of protonation and/or tautomeric states at physiological pH. Intrinsically, standard classical molecular dynamics (MD) simulations cannot describe such equilibria between states, which negatively impacts the prediction of key molecular properties in silico. Following the formalism described by de Oliveira and co-workers (J. Chem. Theory Comput. 2019, 15, 424-435) to consider the influence of all states on the binding process based on alchemical free-energy calculations, we demonstrate in this work that the multistate method replica-exchange enveloping distribution sampling (RE-EDS) is well suited to describe molecules with multiple protonation and/or tautomeric states in a single simulation. We apply our methodology to a series of eight inhibitors of factor Xa with two protonation states and a series of eight inhibitors of glycogen synthase kinase 3β (GSK3β) with two tautomeric states. In particular, we show that given a sufficient phase-space overlap between the states, RE-EDS is computationally more efficient than standard pairwise free-energy methods.

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

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

Development and test of highly accurate endpoint free energy methods. 3: partition coefficient prediction using a Poisson-Boltzmann method combined with a solvent accessible surface area model for SAMPL challenges

Accurately predicting solvation free energy is the key to predict protein-ligand binding free energy. In addition, the partition coefficient (log P), which is an important physicochemical property that determines the distribution of a drug in vivo, can be derived directly from transfer free energies, i.e., the difference between solvation free energies (SFEs) in different solvents. Within the Statistical Assessment of the Modeling of Proteins and Ligands (SAMPL) 9 challenge, we applied the Poisson-Boltzmann (PB) surface area (SA) approach to predict the toluene/water transfer free energy and partition coefficient (log Ptoluene/water) from SFEs. For each solute, only a single conformation automatically generated by the free software Open Babel was used. The PB calculation directly adopts our previously optimized boundary definition - a set of general AMBER force field 2 (GAFF2) atom-type based sphere radii for solute atoms. For the non-polar SA model, we newly developed the solvent-related molecular surface tension parameters γ and offset b for toluene and cyclohexane targeting experimental SFEs. This approach yielded the highest predictive accuracy in terms of root mean square error (RMSE) of 1.52 kcal mol-1 in transfer free energy for 16 small drug molecules among all 18 submissions in the SAMPL9 blind prediction challenge. The re-evaluation of the challenge set using multi-conformation strategies based on molecular dynamics (MD) simulations further reduces the prediction RMSE to 1.33 kcal mol-1. At the same time, an additional evaluation of our PBSA method on the SAMPL5 cyclohexane/water distribution coefficient (log Dcyclohexane/water) prediction revealed that our model outperformed COSMO-RS, the best submission model with RMSEPBSA = 1.88 versus RMSECOSMO-RS = 2.11 log units. Two external log Ptoluene/water and log Pcyclohexane/water datasets that contain 110 and 87 data points, respectively, are collected for extra validation and provide an in-depth insight into the error source of the PBSA method.

Blind prediction of toluene/water partition coefficients using COSMO-RS: results from the SAMPL9 challenge

Accurately predicting partition coefficients log P is crucial for reducing costs and accelerating drug design as it provides valuable information about the bioavailability, pharmacokinetics, and toxicity of different drug candidates. However, the performance of the existing methods is ambiguous, making it unclear whether these methods can be effectively utilized in drug discovery. To assess the performance of these methods, a series of SAMPL challenges have been conducted over the past few years, aiming to enable the development and validation of predictive models. In this study, we present two independent contributions to the SAMPL9 challenge for predicting the toluene/water partition coefficients for 16 molecules. Both submissions, A and B, use the COSMO-RS approach, albeit in slightly different procedures, to compute the transfer free energies from water to toluene of the molecules presented in the challenge, and consequently, their corresponding log P values. Based on the results, COSMO-RS submission A achieves the top position with an R2 value of 0.93 while it ranks second in terms of root-mean-square error (RMSE) with a value of 1.23 log P units. COSMO-RS submission B achieves an R2 value of 0.83 and an RMSE value of 1.48 log P units. Following the challenge, we predict the log P values using a neural network model, which was pre-trained on COSMO-RS data achieving an R2 of 0.92 and RMSE of 1.04 log P units. Compared to previous SAMPL challenges, all contributions displayed large deviations in predicting the toluene/water partition coefficient. These large deviations emphasize that further research is needed to develop accurate and reliable methods for modeling solvent effects on small molecule transfer-free energies.

Advances in Computational Approaches for Estimating Passive Permeability in Drug Discovery

Passive permeation of cellular membranes is a key feature of many therapeutics. The relevance of passive permeability spans all biological systems as they all employ biomembranes for compartmentalization. A variety of computational techniques are currently utilized and under active development to facilitate the characterization of passive permeability. These methods include lipophilicity relations, molecular dynamics simulations, and machine learning, which vary in accuracy, complexity, and computational cost. This review briefly introduces the underlying theories, such as the prominent inhomogeneous solubility diffusion model, and covers a number of recent applications. Various machine-learning applications, which have demonstrated good potential for high-volume, data-driven permeability predictions, are also discussed. Due to the confluence of novel computational methods and next-generation exascale computers, we anticipate an exciting future for computationally driven permeability predictions.

Recent Studies of Artificial Intelligence on In Silico Drug Absorption

Absorption is an important area of research in pharmacochemistry and drug development, because the drug has to be absorbed before any drug effects can occur. Furthermore, the ADMET (Absorption, Distribution, Metabolism, Excretion, and Toxicity) profile of drugs can be directly and considerably altered by modulating factors affecting absorption. Many drugs in development fail because of poor absorption. The research and continuous efforts of researchers in recent years have brought many successes and promises in drug absorption property prediction, especially in silico, which helps to reduce the time and cost significantly for screening undesirable drug candidates. In this report, we explicitly provide an overview of recent in silico studies on predicting absorption properties, especially from 2019 to the present, using artificial intelligence. Additionally, we have collected and investigated public databases that support absorption prediction research. On those grounds, we also proposed the challenges and development directions of absorption prediction in the future. We hope this review can provide researchers with valuable guidelines on absorption prediction to facilitate the development of newer approaches in drug discovery.

Adaptation of Empirical Methods to Predict the LogD of Triazine Macrocycles

Octanol/water partition coefficients guide drug design, but algorithms do not always accurately predict these values. For cationic triazine macrocycles that adopt a conserved folded shape in solution, common algorithms fall short. Here, the logD values for 12 macrocycles differing in amino acid choice were predicted and then measured experimentally. On average, AlogP, XlogP, and ChemAxon predictions deviate by 0.9, 2.8, and 3.9 log units, with XlogP overestimating lipophilicity and AlogP and ChemAxon underestimating lipophilicity. Importantly, however, a linear relationship (R2 > 0.98) exists between the values predicted by AlogP and the experimentally determined logD values, thus enabling more accurate predictions.

LogD7.4 prediction enhanced by transferring knowledge from chromatographic retention time, microscopic pKa and logP

Lipophilicity is a fundamental physical property that significantly affects various aspects of drug behavior, including solubility, permeability, metabolism, distribution, protein binding, and toxicity. Accurate prediction of lipophilicity, measured by the logD7.4 value (the distribution coefficient between n-octanol and buffer at physiological pH 7.4), is crucial for successful drug discovery and design. However, the limited availability of data for logD modeling poses a significant challenge to achieving satisfactory generalization capability. To address this challenge, we have developed a novel logD7.4 prediction model called RTlogD, which leverages knowledge from multiple sources. RTlogD combines pre-training on a chromatographic retention time (RT) dataset since the RT is influenced by lipophilicity. Additionally, microscopic pKa values are incorporated as atomic features, providing valuable insights into ionizable sites and ionization capacity. Furthermore, logP is integrated as an auxiliary task within a multitask learning framework. We conducted ablation studies and presented a detailed analysis, showcasing the effectiveness and interpretability of RT, pKa, and logP in the RTlogD model. Notably, our RTlogD model demonstrated superior performance compared to commonly used algorithms and prediction tools. These results underscore the potential of the RTlogD model to improve the accuracy and generalization of logD prediction in drug discovery and design. In summary, the RTlogD model addresses the challenge of limited data availability in logD modeling by leveraging knowledge from RT, microscopic pKa, and logP. Incorporating these factors enhances the predictive capabilities of our model, and it holds promise for real-world applications in drug discovery and design scenarios.

Bridging the Gap between Differential Mobility, Log S, and Log P Using Machine Learning and SHAP Analysis

Aqueous solubility, log S, and the water-octanol partition coefficient, log P, are physicochemical properties that are used to screen the viability of drug candidates and to estimate mass transport in the environment. In this work, differential mobility spectrometry (DMS) experiments performed in microsolvating environments are used to train machine learning (ML) frameworks that predict the log S and log P of various molecule classes. In lieu of a consistent source of experimentally measured log S and log P values, the OPERA package was used to evaluate the aqueous solubility and hydrophobicity of 333 analytes. With ion mobility/DMS data (e.g., CCS, dispersion curves) as input, we used ML regressors and ensemble stacking to derive relationships with a high degree of explainability, as assessed via SHapley Additive exPlanations (SHAP) analysis. The DMS-based regression models returned scores of R² = 0.67 and RMSE = 1.03 ± 0.10 for log S predictions and R² = 0.67 and RMSE = 1.20 ± 0.10 for log P after 5-fold random cross-validation. SHAP analysis reveals that the regressors strongly weighted gas-phase clustering in log P correlations. The addition of structural descriptors (e.g., # of aromatic carbons) improved log S predictions to yield RMSE = 0.84 ± 0.07 and R² = 0.78. Similarly, log P predictions using the same data resulted in an RMSE of 0.83 ± 0.04 and R² = 0.84. The SHAP analysis of log P models highlights the need for additional experimental parameters describing hydrophobic interactions. These results were achieved with a smaller dataset (333 instances) and minimal structural correlation compared to purely structure-based models, underscoring the value of employing DMS data in predictive models.

Prediction of toluene/water partition coefficients in the SAMPL9 blind challenge: assessment of machine learning and IEF-PCM/MST continuum solvation models

In recent years the use of partition systems other than the widely used biphasic n-octanol/water has received increased attention to gain insight into the molecular features that dictate the lipophilicity of compounds. Thus, the difference between n-octanol/water and toluene/water partition coefficients has proven to be a valuable descriptor to study the propensity of molecules to form intramolecular hydrogen bonds and exhibit chameleon-like properties that modulate solubility and permeability. In this context, this study reports the experimental toluene/water partition coefficients (log P_tol/w) for a series of 16 drugs that were selected as an external test set in the framework of the Statistical Assessment of the Modeling of Proteins and Ligands (SAMPL) blind challenge. This external set has been used by the computational community to calibrate their methods in the current edition (SAMPL9) of this contest. Furthermore, the study also investigates the performance of two computational strategies for the prediction of log P_tol/w. The first relies on the development of two machine learning (ML) models, which are built up by combining the selection of 11 molecular descriptors in conjunction with either the multiple linear regression (MLR) or the random forest regression (RFR) model to target a dataset of 252 experimental log P_tol/w values. The second consists of the parametrization of the IEF-PCM/MST continuum solvation model from B3LYP/6-31G(d) calculations to predict the solvation free energies of 163 compounds in toluene and benzene. The performance of the ML and IEF-PCM/MST models has been calibrated against external test sets, including the compounds that define the SAMPL9 log P_tol/w challenge. The results are used to discuss the merits and weaknesses of the two computational approaches.

COSMO-RS blind prediction of distribution coefficients and aqueous pKa values from the SAMPL8 challenge

The SAMPL8 blind prediction challenge, which addresses the acid/base dissociation constants (pKa) and the distribution coefficients (logD), was addressed by the Conductor like Screening Model for Realistic Solvation (COSMO-RS). Using the COSMOtherm implementation of COSMO-RS together with a rigorous conformational sampling, yielded logD predictions with a root mean square deviation (RMSD) of 1.36 log units over all 11 compounds and seven bi-phasic systems of the data set, which was the most accurate of all contest submissions (logD).For the SAMPL8 pKa competition, participants were asked to report the standard state free energies of all microstates, which were then used to calculate the macroscopic pKa. We have used COSMO-RS based linear free energy fit models to calculate the requested energies. The assignment of the calculated and experimental pKa values was made on the basis of the popular transitions, i.e. the transition hat was predicted by the majority of the submissions. With this assignment and a model that covers both, pKa and base pKa, we achieved an RMSD of 3.44 log units (18 pKa values of 14 molecules), which is the second place of the six ranked submissions. By changing to an assignment that is based on the experimental transition curves, the RMSD reduces to 1.65. In addition to the ranked contribution, we submitted two more data sets, one for the standard pKa model and one or the standard base pKa model of COSMOtherm. Using the experiment based assignment with the predictions of the two sets we received a RMSD of 1.42 log units (25 pKa values of 20 molecules). The deviation mainly arises from a single outlier compound, the omission of which leads to an RMSD of 0.89 log units.

A computational and laboratory approach for the investigation of interactions of peptide conjugated natural terpenes with EpHA2 receptor

CONTEXT

Ephrin type A receptor 2 (EphA2) is a well-known drug target for cancer treatment due to its overexpression in numerous types of cancers. Thus, it is crucial to determine the binding interactions of this receptor with both the ligand-binding domain (LBD) and the kinase-binding domain (KBD) through a targeted approach in order to modulate its activity. In this work, natural terpenes with inherent anticancer properties were conjugated with short peptides YSAYP and SWLAY that are known to bind to the LBD of EphA2 receptor. We examined the binding interactions of six terpenes (maslinic acid, levopimaric acid, quinopimaric acid, oleanolic, polyalthic, and hydroxybetulinic acid) conjugated to the above peptides with the ligand-binding domain (LBD) of EphA2 receptor computationally. Additionally, following the "target-hopping approach," we also examined the interactions of the conjugates with the KBD. Our results indicated that most of the conjugates showed higher binding interactions with the EphA2 kinase domain compared to LBD. Furthermore, the binding affinities of the terpenes increased upon conjugating the peptides with the terpenes. In order to further investigate the specificity toward EphA2 kinase domain, we also examined the binding interactions of the terpenes conjugated to VPWXE (x = norleucine), as VPWXE has been shown to bind to other RTKs. Our results indicated that the terpenes conjugated to SWLAY in particular showed high efficacy toward binding to the KBD. We also designed conjugates where in the peptide portion and the terpenes were separated by a butyl (C4) group linker to examine if the binding interactions could be enhanced. Docking studies showed that the conjugates with linkers had enhanced binding with the LBD compared to those without linkers, though binding remained slightly higher without linkers toward the KBD. As a proof of concept, maslinate and oleanolate conjugates of each of the peptides were then tested with F98 tumor cells which are known to overexpress EphA2 receptor. Results indicated that the oleanolate-amido-SWLAY conjugates were efficacious in reducing the cell proliferation of the tumor cells and may be potentially developed and further studied for targeting tumor cells overexpressing the EphA2 receptor. To test if these conjugates could bind to the receptor and potentially function as kinase inhibitors, we conducted SPR analysis and ADP-Glo assay. Our results indicated that OA conjugate with SWLAY showed the highest inhibition.

METHODS

Docking studies were carried out using AutoDock Vina, v.1.2.0; Molecular Dynamics and MMGBSA calculations were carried out through Schrodinger Software DESMOND.

Rapid identification of in vitro cell toxicity using an electrochemical membrane screening platform

This study compares the performance and output of an electrochemical phospholipid membrane platform against respective in vitro cell-based toxicity testing methods using three toxicants of different biological action (chlorpromazine (CPZ), colchicine (COL) and methyl methanesulphonate (MMS)). Human cell lines from seven different tissues (lung, liver, kidney, placenta, intestine, immune system) were used to validate this physicochemical testing system. For the cell-based systems, the effective concentration at 50 % cell death (EC₅₀) values are calculated. For the membrane sensor, a limit of detection (LoD) value was extracted as a quantitative parameter describing the minimum concentration of toxicant which significantly affects the structure of the phospholipid sensor membrane layer. LoD values were found to align well with the EC₅₀ values when acute cell viability was used as an end-point and showed a similar toxicity ranking of the tested toxicants. Using the colony forming efficiency (CFE) or DNA damage as end-point, a different order of toxicity ranking was observed. The results of this study showed that the electrochemical membrane sensor generates a parameter relating to biomembrane damage, which is the predominant factor in decreasing cell viability when in vitro models are acutely exposed to toxicants. These results lead the way to using electrochemical membrane-based sensors for rapid relevant preliminary toxicity screens.

Machine Learning for Fast, Quantum Mechanics-Based Approximation of Drug Lipophilicity

Lipophilicity, as measured by the partition coefficient between octanol and water (log P), is a key parameter in early drug discovery research. However, measuring log P experimentally is difficult for specific compounds and log P ranges. The resulting lack of reliable experimental data impedes development of accurate in silico models for such compounds. In certain discovery projects at Novartis focused on such compounds, a quantum mechanics (QM)-based tool for log P estimation has emerged as a valuable supplement to experimental measurements and as a preferred alternative to existing empirical models. However, this QM-based approach incurs a substantial computational cost, limiting its applicability to small series and prohibiting quick, interactive ideation. This work explores a set of machine learning models (Random Forest, Lasso, XGBoost, Chemprop, and Chemprop3D) to learn calculated log P values on both a public data set and an in-house data set to obtain a computationally affordable, QM-based estimation of drug lipophilicity. The message-passing neural network model Chemprop emerged as the best performing model with mean absolute errors of 0.44 and 0.34 log units for scaffold split test sets of the public and in-house data sets, respectively. Analysis of learning curves suggests that a further decrease in the test set error can be achieved by increasing the training set size. While models directly trained on experimental data perform better at approximating experimentally determined log P values than models trained on calculated values, we discuss the potential advantages of using calculated log P values going beyond the limits of experimental quantitation. We analyze the impact of the data set splitting strategy and gain insights into model failure modes. Potential use cases for the presented models include pre-screening of large compound collections and prioritization of compounds for full QM calculations.

Simultaneous effect of different chromatographic conditions on the chromatographic retention of pentapeptide derivatives (HGRFG and NPNPT)

Introduction: Oligopeptides exhibit great prospects for clinical application and its separation is of great importance in new drug development. Methods: To accurately predict the retention of pentapeptides with analogous structures in chromatography, the retention times of 57 pentapeptide derivatives in seven buffers at three temperatures and four mobile phase compositions were measured via reversed-phase high-performance liquid chromatography. The parameters ( k H A , k A , and p K a ) of the acid-base equilibrium were obtained by fitting the data corresponding to a sigmoidal function. We then studied the dependence of these parameters on the temperature (T), organic modifier composition (φ, methanol volume fraction), and polarity ( P m N parameter). Finally, we proposed two six-parameter models with (1) pH and T and (2) pH and φ or P m N as the independent variables. These models were validated for their prediction capacities by linearly fitting the predicted retention factor k-value and the experimental k-value. Results: The results showed that log k H A and log k A exhibited linear relationships with 1 / T , φ or P m N for all pentapeptides, especially for the acid pentapeptides. In the model of pH and T, the correlation coefficient (R²) of the acid pentapeptides was 0.8603, suggesting a certain prediction capability of chromatographic retention. Moreover, in the model of pH and φ or P m N , the R² values of the acid and neutral pentapeptides were greater than 0.93, and the average root mean squared error was approximately 0.3, indicating that the k-values could be effectively predicted. Discussion: In summary, the two six-parameter models were appropriate to characterize the chromatographic retention of amphoteric compounds, especially the acid or neutral pentapeptides, and could predict the chromatographic retention of pentapeptide compounds.

Machine learning methods for pKa prediction of small molecules: Advances and challenges

The acid-base dissociation constant (pK_a) is a fundamental property influencing many ADMET properties of small molecules. However, rapid and accurate pK_a prediction remains a great challenge. In this review, we outline the current advances in machine-learning-based QSAR models for pK_a prediction, including descriptor-based and graph-based approaches, and summarize their pros and cons. Moreover, we highlight the current challenges and future directions regarding experimental data, crucial factors influencing pK_a and in silico prediction tools. We hope that this review can provide a practical guidance for the follow-up studies.

Predicting octanol/water partition coefficients and pKa for the SAMPL7 challenge using the SM12, SM8 and SMD solvation models

Blind predictions of octanol/water partition coefficients and pKa at 298.15 K for 22 drug-like compounds were made for the SAMPL7 challenge. Octanol/water partition coefficients were predicted from solvation free energies computed using electronic structure calculations with the SM12, SM8 and SMD solvation models. Within these calculations we compared the use of gas- and solution-phase optimized geometries of the solute. Based on these calculations we found that in general the use of solution phase-optimized geometries increases the affinity of the solutes for water as compared to octanol, with the use of gas-phase optimized geometries resulting in the better agreement with experiment. The pKa is computed using the direct approach, scaled solvent-accessible surface model, and the inclusion of an explicit water molecule, where the latter two methods have previously been shown to offer improved predictions as compared to the direct approach. We find that the use of an explicit water molecule provides superior predictions, and that the predicted macroscopic pKa is sensitive to the employed microstates.

Lipophilicity prediction of three photosensitizers by liquid-liquid extraction, HPLC, and DFT methods

Photodynamic therapy (PDT) is a method of treating precancerous diseases and malignant neoplasms. The efficacy of PDT depends on different parameters such as light dosimetry, oxygen availability, and photophysical and physical-chemical properties of the photosensitizer. In PDT, a photosensitizer is activated using light to promote oxygen photosensitization and cellular transport plays a key role in the reach of it to the desired tissue. In particular, to know the effectiveness of the drug delivery in PDT and its dosage forms to target damaged organs, along with such characteristics as water solubility, it is important to know the ability of a substance to penetrate through cell membrane or accumulate in it. Lipophilicity is used to quantify the earlier-described abilities. We evaluated the lipophilicity of three selected photosensitizers (PS) (protoporphyrin IX, pyropheophorbide-a and photofrin) by means of three different methods: octanol-water distribution methods (shake-flask), reversed-phase high-performance liquid chromatography (HPLC) and theoretical calculations based on density functional theory (DFT). We describe and compare the results of these various methods.

Drug Disposition in Subjects with Obesity: The Research Work of Darrell R. Abernethy

In 1979, the late Dr. Darrell R. Abernethy and colleagues began a series of clinical studies aimed at understanding the pertinent determinants of drug distribution, elimination, and clearance in obesity, and how those variables are interconnected. The studies confirmed that volume of distribution (Vd) and clearance are the principal independent biological variables, which conjointly determine elimination half-life as a dependent variable. For drugs distributed by passive diffusion, their pharmacokinetic Vd - after correcting for plasma protein binding - was increased in obesity, depending in part on the physicochemical lipophilicity of the individual drugs, and the quantitative extent of obesity in overweight individuals. Across all studies, the ratio of mean clearance in obese divided by control groups had an overall median value of 1.21 (range: 0.75 to 3.11), indicating a small and variable effect of obesity on clearance, without clear directionality. Since drug clearance was not clearly related to lipophilicity or degree of obesity, the prolonged half-life of lipophilic drugs in obese patients was largely explained by the increased Vd. Dr. Abernethy further identified delayed attainment of steady-state after initiation of multiple-dose treatment, and delayed washout after termination of dosage, as potential clinical consequences of the extended half-life in obese persons. These consequences for specific drugs have been recently emphasized in contemporary studies of chronic dosage in subjects with obesity. Without data identifying an obesity-related change in clearance for a specific drug, maintenance doses (in milligrams) should be based on ideal weight rather than adjusted upward based on total weight. This article is protected by copyright. All rights reserved.

Improving Small Molecule pKa Prediction Using Transfer Learning With Graph Neural Networks

Enumerating protonation states and calculating microstate pK_a values of small molecules is an important yet challenging task for lead optimization and molecular modeling. Commercial and non-commercial solutions have notable limitations such as restrictive and expensive licenses, high CPU/GPU hour requirements, or the need for expert knowledge to set up and use. We present a graph neural network model that is trained on 714,906 calculated microstate pK_a predictions from molecules obtained from the ChEMBL database. The model is fine-tuned on a set of 5,994 experimental pK_a values significantly improving its performance on two challenging test sets. Combining the graph neural network model with Dimorphite-DL, an open-source program for enumerating ionization states, we have developed the open-source Python package pkasolver, which is able to generate and enumerate protonation states and calculate pK_a values with high accuracy.

Implementation and Optimization of the Embedded Cluster Reference Interaction Site Model with Atomic Charges

In this work, we implemented the embedded cluster reference interaction site model (EC-RISM) originally developed by Kloss, Heil, and Kast (J. Phys. Chem. B2008, 112, 4337–4343). This method combines quantum mechanical calculations with the 3D reference interaction site model (3D-RISM). Numerous options, such as buffer, grid space, basis set, charge model, water model, closure relation, and so forth, were investigated to find the best settings. Additionally, the small point charges, which are derived from the solvent distribution from the 3D-RISM solution to represent the solvent in the QM calculation, were neglected to reduce the overhead without the loss of accuracy. On the MNSOL[a], MNSOL, and FreeSolv databases, our implemented and optimized method provides solvation free energies in water with 5.70, 6.32, and 6.44 kJ/mol root-mean-square deviations, respectively, but with different settings, 5.22, 6.08, and 6.63 kJ/mol can also be achieved. Only solvent models containing fitting parameters, like COSMO-RS and EC-RISM with universal correction and directly used electrostatic potential, perform better than our EC-RISM implementation with atomic charges.

SAMPL7 protein-ligand challenge: A community-wide evaluation of computational methods against fragment screening and pose-prediction

A novel crystallographic fragment screening data set was generated and used in the SAMPL7 challenge for protein-ligands. The SAMPL challenges prospectively assess the predictive power of methods involved in computer-aided drug design. Application of various methods to fragment molecules are now widely used in the search for new drugs. However, there is little in the way of systematic validation specifically for fragment-based approaches. We have performed a large crystallographic high-throughput fragment screen against the therapeutically relevant second bromodomain of the Pleckstrin-homology domain interacting protein (PHIP2) that revealed 52 different fragments bound across 4 distinct sites, 47 of which were bound to the pharmacologically relevant acetylated lysine (Kac) binding site. These data were used to assess computational screening, binding pose prediction and follow-up enumeration. All submissions performed randomly for screening. Pose prediction success rates (defined as less than 2 Å root mean squared deviation against heavy atom crystal positions) ranged between 0 and 25% and only a very few follow-up compounds were deemed viable candidates from a medicinal-chemistry perspective based on a common molecular descriptors analysis. The tight deadlines imposed during the challenge led to a small number of submissions suggesting that the accuracy of rapidly responsive workflows remains limited. In addition, the application of these methods to reproduce crystallographic fragment data still appears to be very challenging. The results show that there is room for improvement in the development of computational tools particularly when applied to fragment-based drug design.

Preformulation Studies of Thymopentin: Analytical Method Development, Physicochemical Properties, Kinetic Degradation Investigations and Formulation Perspective

Thymopentin (TP5) is a synthetic pentapeptide with immunomodulatory properties. Given the previously described poor absorption of TP5, preformulation data is required to support effective formulation development. In this manuscript, an analytical method of TP5 was developed and validated to determine the aqueous solubility, stability, and Log P of TP5. Thermal properties were investigated, and chemical, physical and enzymatic degradation were evaluated. TP5 was informed to load in a microemulsion (ME) system according to the preformulation parameters and characterized for rheological behavior, droplet size, morphology and in vitro drug release. TP5 displayed high aqueous solubility (294.3 mg/mL), low Log P (-4.2) and 2% water content with a melting temperature of 193 °C. TP5 degraded rapidly in alkaline conditions, at elevated temperature, in oxidizing agents, and with UV exposure, however TP5 had a longer half-life in acidic conditions. The fastest enzymatic degradation was with Trypsin (half-life 6.3 hours) compared with other digestive enzymes. The different degradation pathways followed first-order kinetics, and half-lives were obtained from the kinetic studies. The TP5 loaded ME exhibited a droplet size of 143 ± 35 nm with a Higuchi-model fitted sustained release profile for 24 hours. These data justify and support the design of formulations to stabilize and enhance the absorption of TP5, with a ME formulation demonstrated.

Chemical reactivity prediction: current methods and different application areas

The ability to predict chemical reactivity of a molecule is highly desirable in drug discovery, both ex vivo (synthetic route planning, formulation, stability) and in vivo: metabolic reactions determine pharmacodynamics, pharmacokinetics and potential toxic effects, and early assessment of liabilities is vital to reduce attrition rates in later stages of development. Quantum mechanics offer a precise description of the interactions between electrons and orbitals in the breaking and forming of new bonds. Modern algorithms and faster computers have allowed the study of more complex systems in a punctual and accurate fashion, and answers for chemical questions around stability and reactivity can now be provided. Through machine learning, predictive models can be built out of descriptors derived from quantum mechanics and cheminformatics, even in the absence of experimental data to train on. In this article, current progress on computational reactivity prediction is reviewed: applications to problems in drug design, such as modelling of metabolism and covalent inhibition, are highlighted and unmet challenges are posed.

Predicting octanol/water partition coefficients using molecular simulation for the SAMPL7 challenge: comparing the use of neat and water saturated 1-octanol

Blind predictions of octanol/water partition coefficients at 298.15 K for 22 drug-like compounds were made for the SAMPL7 challenge. The octanol/water partition coefficients were predicted using solvation free energies computed using molecular dynamics simulations, wherein we considered the use of both pure and water-saturated 1-octanol to model the octanol-rich phase. Water and 1-octanol were modeled using TIP4P and TrAPPE-UA, respectively, which have been shown to well reproduce the experimental mutual solubility, and the solutes were modeled using GAFF. After the close of the SAMPL7 challenge, we additionally made predictions using TIP4P/2005 water. We found that the predictions were sensitive to the choice of water force field. However, the effect of water in the octanol-rich phase was found to be even more significant and non-negligible. The effect of inclusion of water was additionally sensitive to the chemical structure of the solute.

SAMPL7 physical property prediction from EC-RISM theory

Inspired by the successful application of the embedded cluster reference interaction site model (EC-RISM), a combination of quantum–mechanical calculations with three-dimensional RISM theory to predict Gibbs energies of species in solution within the SAMPL6.1 (acidity constants, pK_a) and SAMPL6.2 (octanol–water partition coefficients, log P) the methodology was applied to the recent SAMPL7 physical property challenge on aqueous pK_a and octanol–water log P values. Not part of the challenge but provided by the organizers, we also computed distribution coefficients log D_7.4 from predicted pK_a and log P data. While macroscopic pK_a predictions compared very favorably with experimental data (root mean square error, RMSE 0.72 pK units), the performance of the log P model (RMSE 1.84) fell behind expectations from the SAMPL6.2 challenge, leading to reasonable log D_7.4 predictions (RMSE 1.69) from combining the independent calculations. In the post-submission phase, conformations generated by different methodology yielded results that did not significantly improve the original predictions. While overall satisfactory compared to previous log D challenges, the predicted data suggest that further effort is needed for optimizing the robustness of the partition coefficient model within EC-RISM calculations and for shaping the agreement between experimental conditions and the corresponding model description.

Multitask machine learning models for predicting lipophilicity (logP) in the SAMPL7 challenge

Accurate prediction of lipophilicity—logP—based on molecular structures is a well-established field. Predictions of logP are often used to drive forward drug discovery projects. Driven by the SAMPL7 challenge, in this manuscript we describe the steps that were taken to construct a novel machine learning model that can predict and generalize well. This model is based on the recently described Directed-Message Passing Neural Networks (D-MPNNs). Further enhancements included: both the inclusion of additional datasets from ChEMBL (RMSE improvement of 0.03), and the addition of helper tasks (RMSE improvement of 0.04). To the best of our knowledge, the concept of adding predictions from other models (Simulations Plus logP and logD@pH7.4, respectively) as helper tasks is novel and could be applied in a broader context. The final model that we constructed and used to participate in the challenge ranked 2/17 ranked submissions with an RMSE of 0.66, and an MAE of 0.48 (submission: Chemprop). On other datasets the model also works well, especially retrospectively applied to the SAMPL6 challenge where it would have ranked number one out of all submissions (RMSE of 0.35). Despite the fact that our model works well, we conclude with suggestions that are expected to improve the model even further.

Prediction of n-octanol/water partition coefficients and acidity constants (pKa) in the SAMPL7 blind challenge with the IEFPCM-MST model

Within the scope of SAMPL7 challenge for predicting physical properties, the Integral Equation Formalism of the Miertus-Scrocco-Tomasi (IEFPCM/MST) continuum solvation model has been used for the blind prediction of n-octanol/water partition coefficients and acidity constants of a set of 22 and 20 sulfonamide-containing compounds, respectively. The log P and pK_a were computed using the B3LPYP/6-31G(d) parametrized version of the IEFPCM/MST model. The performance of our method for partition coefficients yielded a root-mean square error of 1.03 (log P units), placing this method among the most accurate theoretical approaches in the comparison with both globally (rank 8th) and physical (rank 2nd) methods. On the other hand, the deviation between predicted and experimental pK_a values was 1.32 log units, obtaining the second best-ranked submission. Though this highlights the reliability of the IEFPCM/MST model for predicting the partitioning and the acid dissociation constant of drug-like compounds compound, the results are discussed to identify potential weaknesses and improve the performance of the method.

Precise force-field-based calculations of octanol-water partition coefficients for the SAMPL7 molecules

We predicted water-octanol partition coefficients for the molecules in the SAMPL7 challenge with explicit solvent classical molecular dynamics (MD) simulations. Water hydration free energies and octanol solvation free energies were calculated with a windowed alchemical free energy approach. Three commonly used force fields (AMBER GAFF, CHARMM CGenFF, OPLS-AA) were tested. Special emphasis was placed on converging all simulations, using a criterion developed for the SAMPL6 challenge. In aggregate, over 1000 [Formula: see text]s of simulations were performed, with some free energy windows remaining not fully converged even after 1 [Formula: see text]s of simulation time. Nevertheless, the amount of sampling produced [Formula: see text] estimates with a precision of 0.1 log units or better for converged simulations. Despite being probably as fully sampled as can expected and is feasible, the agreement with experiment remained modest for all force fields, with no force field performing better than 1.6 in root mean squared error. Overall, our results indicate that a large amount of sampling is necessary to produce precise [Formula: see text] predictions for the SAMPL7 compounds and that high precision does not necessarily lead to high accuracy. Thus, fundamental problems remain to be solved for physics-based [Formula: see text] predictions.