Application of Artificial Neural Networks to Predict the Intrinsic Solubility of Drug-Like Molecules

Machine learning (ML) approaches are receiving increasing attention from pharmaceutical companies and regulatory agencies, given their ability to mine knowledge from available data. In drug discovery, for example, they are employed in quantitative structure–property relationship (QSPR) models to predict biological properties from the chemical structure of a drug molecule. In this paper, following the Second Solubility Challenge (SC-2), a QSPR model based on artificial neural networks (ANNs) was built to predict the intrinsic solubility (logS0) of the 100-compound low-variance tight set and the 32-compound high-variance loose set provided by SC-2 as test datasets. First, a training dataset of 270 drug-like molecules with logS0 value experimentally determined was gathered from the literature. Then, a standard three-layer feed-forward neural network was defined by using 10 ChemGPS physico-chemical descriptors as input features. The developed ANN showed adequate predictive performances on both of the SC-2 test datasets. Benefits and limitations of ML approaches have been highlighted and discussed, starting from this case-study. The main findings confirmed that ML approaches are an attractive and promising tool to predict logS0; however, many aspects, such as data quality, molecular descriptor computation and selection, and assessment of applicability domain, are crucial but often neglected, and should be carefully considered to improve predictions based on ML.

Physicochemical and Biopharmaceutical Evaluation of Casiopeina III-Ea, a New Compound with Antineoplastic Activity

In the present work, the physicochemical (pka, intrinsic solubility, Log D distribution coefficient) and biopharmaceutical characterization (IC50 studies in MDCK cells) of Casiopeina III-Ea, a Cu (II) mixed chelate compound, was carried out. These parameters will tell us about the behaviour of Casiopeina III-Ea, as a new antineoplastic agent, in physiological media. This basic research of the Faculty of Chemistry of the UNAM, is directed towards the maturation of a technology, which consists of the pharmaceutical development of an antineoplastic copper coordination compound Casiopeina III-Ea. Resumen. En el presente trabajo se realizó la caracterización fisicoquímica (pka, solubilidad intrínseca, coeficiente de distribución Log D) y biofarmacéutica (estudios IC50 en células MDCK) de Casiopeina III-Ea, un nuevo compuesto quelato mixto de cobre (II). Estos parámetros nos dirán mucho sobre el comportamiento de Casiopeina III-Ea, como un nuevo agente antineoplásico, en medios fisiológicos. Esta investigación básica de la Facultad de Química de la UNAM, está dirigida hacia la maduración de una tecnología, que consiste en el desarrollo farmacéutico de un compuesto de coordinación antineoplásico de cobre Casiopeina III-Ea.

Surface Engineering of Fenofibrate Nanocrystals Using Nano-by-Design Multivariate Integration: A Biopharmaceutical and Pharmacokinetic Perspective

Introduction: Surface engineering of nanocrystals for improving the biopharmaceutical features is a multivariate process involving numerous formulation and process variables, thus making it a complicated process to get the desired biopharmaceutical quality profile. Nano-by-design is hereby proposed as an approach to nanonize an orally active, lipid lowering fenofibrate, to improve feasibility in product development. Methodology: Top-down wet ball milling (media milling) in zirconia planetary chamber was methodically explored for improving the solubility and bioavailability of fenofibrate by formulating a nanosuspension using polyvinyl alcohol as a stabilizer. Several influencing variables were screened using a systematic one-factor-at-a-time approach. DSC, SEM, XRD, and FTIR were utilized for physical characterization of the product during the development stage and study the effect of milling time, milling speed, fenofibrate: stabilizer ratio, premilling time and stabilizer concentration. Potential risk factors affecting critical quality biopharmaceutical attributes of fenofibrate nanocrystals like size, zeta potential, in vitro release, crystallinity and intrinsic solubility were optimized to improve pharmacokinetic performance. Result: Formulated nanosized fenofibrate exhibited a crystalize nature as evident from XRD and DSC, 411 nm size, and a rapid but complete dissolution (~99% in 30 min). This resulted into quick onset of action and improved bioavailability as observed from 51.46% shorter Tmax, 82.63% higher Cmax, and 69.34% higher AUC0–24h, respectively.

Bottom-Up Physiologically Based Oral Absorption Modeling of Free Weak Base Drugs

In this study, we systematically evaluated “bottom-up” physiologically based oral absorption modeling, focusing on free weak base drugs. The gastrointestinal unified theoretical framework (the GUT framework) was employed as a simple and transparent model. The oral absorption of poorly soluble free weak base drugs is affected by gastric pH. Alternation of bulk and solid surface pH by dissolving drug substances was considered in the model. Simple physicochemical properties such as pKa, the intrinsic solubility, and the bile micelle partition coefficient were used as input parameters. The fraction of a dose absorbed (Fa) in vivo was obtained by reanalyzing the pharmacokinetic data in the literature (15 drugs, a total of 85 Fa data). The AUC ratio with/without a gastric acid-reducing agent (AUCr) was collected from the literature (22 data). When gastric dissolution was neglected, Fa was underestimated (absolute average fold error (AAFE) = 1.85, average fold error (AFE) = 0.64). By considering gastric dissolution, predictability was improved (AAFE = 1.40, AFE = 1.04). AUCr was also appropriately predicted (AAFE = 1.54, AFE = 1.04). The Fa values of several drugs were slightly overestimated (less than 1.7-fold), probably due to neglecting particle growth in the small intestine. This modeling strategy will be of great importance for drug discovery and development.

Prediction of aqueous intrinsic solubility of druglike molecules using Random Forest regression trained with Wiki-pS0 database

The accurate prediction of solubility of drugs is still problematic. It was thought for a long time that shortfalls had been due the lack of high-quality solubility data from the chemical space of drugs. This study considers the quality of solubility data, particularly of ionizable drugs. A database is described, comprising 6355 entries of intrinsic solubility for 3014 different molecules, drawing on 1325 citations. In an earlier publication, many factors affecting the quality of the measurement had been discussed, and suggestions were offered to improve ways of extracting more reliable information from legacy data. Many of the suggestions have been implemented in this study. By correcting solubility for ionization (i.e., deriving intrinsic solubility, S0) and by normalizing temperature (by transforming measurements performed in the range 10-50 °C to 25 °C), it can now be estimated that the average interlaboratory reproducibility is 0.17 log unit. Empirical methods to predict solubility at best have hovered around the root mean square error (RMSE) of 0.6 log unit. Three prediction methods are compared here: (a) Yalkowsky’s general solubility equation (GSE), (b) Abraham solvation equation (ABSOLV), and (c) Random Forest regression (RFR) statistical machine learning. The latter two methods were trained using the new database. The RFR method outperforms the other two models, as anticipated. However, the ability to predict the solubility of drugs to the level of the quality of data is still out of reach. The data quality is not the limiting factor in prediction. The statistical machine learning methodologies are probably up to the task. Possibly what’s missing are solubility data from a few sparsely-covered chemical space of drugs (particularly of research compounds). Also, new descriptors which can better differentiate the factors affecting solubility between molecules could be critical for narrowing the gap between the accuracy of the prediction models and that of the experimental data.

Proteome-wide observation of the phenomenon of life on the edge of solubility

To function effectively proteins must avoid aberrant aggregation, and hence they are expected to be expressed at concentrations safely below their solubility limits. By analyzing proteome-wide mass spectrometry data of Caenorhabditis elegans, however, we show that the levels of about three-quarters of the nearly 4,000 proteins analyzed in adult animals are close to their intrinsic solubility limits, indeed exceeding them by about 10% on average. We next asked how aging and functional self-assembly influence these solubility limits. We found that despite the fact that the total quantity of proteins within the cellular environment remains approximately constant during aging, protein aggregation sharply increases between days 6 and 12 of adulthood, after the worms have reproduced, as individual proteins lose their stoichiometric balances and the cellular machinery that maintains solubility undergoes functional decline. These findings reveal that these proteins are highly prone to undergoing concentration-dependent phase separation, which on aging is rationalized in a decrease of their effective solubilities, in particular for proteins associated with translation, growth, reproduction, and the chaperone system.

The Solubility-Permeability Interplay for Solubility-Enabling Oral Formulations

The Biopharmaceutical classification system (BCS) classifies the drugs based on their intrinsic solubility and intestinal permeability. The drugs with good solubility and intestinal permeability have good bioavailability. The drugs with poor solubility and poor permeability have solubility dependent and permeability dependent bioavailability, respectively. In the current pharmaceutical field, most of the drugs have poor solubility. To solve the problem of poor solubility, various solubility enhancement approaches have been successfully used. The effects of these solubility enhancing approaches on the intestinal permeability of the drugs are a matter of concern, and must not be overlooked. The current review article focuses on the effect of various solubility enhancing approaches viz. cyclodextrin, surfactant, cosolvent, hydrotropes, and amorphous solid dispersion, on the intestinal permeability of drugs. This article will help in the designing of the optimized formulations having balanced solubility enhancement without affecting the permeability of drugs.

Multi-lab intrinsic solubility measurement reproducibility in CheqSol and shake-flask methods

This commentary compares 233 CheqSol intrinsic solubility values (log S0) reported in the Wiki-pS0 database for 145 different druglike molecules to the 838 log S0 values determined mostly by the saturation shake-flask (SSF) method for 124 of the molecules from the CheqSol set. The range of log S0 spans from -1.0 to -10.6 (log molar units), averaging at -3.8. The correlation plot between the two methods indicates r2 = 0.96, RMSE = 0.34 log unit, and a slight bias of -0.07 log unit. The average interlaboratory standard deviation (SDi) is slightly better for the CheqSol set than that of the SSF set: SDiCS = 0.15 and SDiSSF = 0.24. The intralaboratory errors reported in the CheqSol method (0.05 log) need to be multiplied by a factor of 3 to match the expected interlaboratory errors for the method. The scale factor, in part, relates to the hidden systematic errors in the single-lab values. It is expected that improved standardizations in the ‘gold standard’ SSF method, as suggested in the recent ‘white paper’ on solubility measurement methodology, should make the SDi of both methods be about ~0.15 log unit. The multi-lab averaged log S0 (and the corresponding SDi) values could be helpful additions to existing training-set molecules used to predict the intrinsic solubility of drugs and druglike molecules.

Aqueous Drug Solubility: What Do We Measure, Calculate and QSPR Predict?

Detailed critical analysis of publications devoted to QSPR of aqueous solubility is presented in the review with discussion of four types of aqueous solubility (three different thermodynamic solubilities with unknown solute structure, intrinsic solubility, solubility in physiological media at pH=7.4 and kinetic solubility), variety of molecular descriptors (from topological to quantum chemical), traditional statistical and machine learning methods as well as original QSPR models.