Lead-like, drug-like or "Pub-like": how different are they?

Academic and industrial research continues to be focused on discovering new classes of compounds based on HTS. Post-HTS analyses need to prioritize compounds that are progressed to chemical probe or lead status. We report trends in probe, lead and drug discovery by examining the following categories of compounds: 385 leads and the 541 drugs that emerged from them; "active" (152) and "inactive" (1488) compounds from the Molecular Libraries Initiative Small Molecule Repository (MLSMR) tested by HTS; "active" (46) and "inactive" (72) compounds from Nature Chemical Biology (NCB) tested by HTS; compounds in the drug development phase (I, II, III and launched), as indexed in MDDR; and medicinal chemistry compounds from WOMBAT, separated into high-activity (5,784 compounds with nanomolar activity or better) and low-activity (30,690 with micromolar activity or less). We examined Molecular weight (MW), molecular complexity, flexibility, the number of hydrogen bond donors and acceptors, LogP-the octanol/water partition coefficient estimated by ClogP and ALOGPS), LogSw (intrinsic water solubility, estimated by ALOGPS) and the number of Rule of five (Ro5) criteria violations. Based on the 50% and 90% distribution moments of the above properties, there were no significant difference between leads of known drugs and "actives" from MLSMR or NCB (chemical probes). "Inactives" from NCB and MLSMR were also found to exhibit similar properties. From these combined sets, we conclude that "Actives" (569 compounds) are less complex, less flexible, and more soluble than drugs (1,651 drugs), and significantly smaller, less complex, less hydrophobic and more soluble than the 5,784 high-activity WOMBAT compounds. These trends indicate that chemical probes are similar to leads with respect to some properties, e.g., complexity, solubility, and hydrophobicity.

Skin permeation rate as a function of chemical structure.

Multilinear and nonlinear QSAR models were built for the skin permeation rate (Log K(p)) of a set of 143 diverse compounds. Satisfactory models were obtained by three approaches applied: (i) CODESSA PRO, (ii) Neural Network modeling using large pools of theoretical molecular descriptors, and (iii) ISIDA modeling based on fragment descriptors. The predictive abilities of the models were assessed by internal and external validations. The descriptors involved in the equations are discussed from the physicochemical point of view to illuminate the factors that influence skin permeation.

Correlation of blood-brain penetration using structural descriptors.

Experimental blood-brain partition coefficients (logBB) for a diverse set of 113 drug molecules are correlated with computed structural descriptors using CODESSA-PRO and ISIDA programs to give statistically significant QSAR models based respectively, on molecular and on fragment descriptors. The linear correlation CODESSA-PRO five-descriptor model has correlation coefficient R2=0.781 and standard deviation s2=0.123. The 'consensus model' of ISIDA gave R2=0.872 and s2=0.047. The developed models were successfully validated using the central nervous system activity data of an external test set of 40 drug molecules.

QSAR modeling of the inhibition of glycogen synthase kinase-3.

Quantitative structure-activity relationship (QSAR) models of the biological activity (pIC50) of 277 inhibitors of Glycogen Synthase Kinase-3 (GSK-3) are developed using geometrical, topological, quantum mechanical, and electronic descriptors calculated by CODESSA PRO. The linear (multilinear regression) and nonlinear (artificial neural network) models obtained link the structures to their reported activity pIC50. The results are discussed in the light of the main factors that influence the inhibitory activity of the GSK-3 enzyme.

Antimalarial activity: a QSAR modeling using CODESSA PRO software.

A quantitative structure-activity relationship (QSAR) modeling of the antimalarial activity of two diverse sets of compounds for each of two strains D6 and NF54 of Plasmodium falciparum is presented. The molecular structural features of compounds are presented by molecular descriptors (geometrical, topological, quantum mechanical, and electronic) calculated using the CODESSA PRO software. Satisfactory multilinear regression models were obtained for data sets of the D6 and NF54 strains, with R2 = 0.84 and 0.89, respectively. The models were also satisfactorily validated internally. The descriptors involved in these equations were related to the mechanism of antimalarial protection.

Integration of virtual and physical screening.

High-throughput screening (HTS) represents the dominant technique for the identification of new lead compounds in current drug discovery. It consists of physical screening (PS) of large libraries of chemicals against one or more specific biological targets. Virtual screening (VS) is a strategy for in silico evaluation of chemical libraries for a given target, and can be integrated to focus the PS process. The present work addresses the integration of both PS and VS, respectively.

QSAR modeling of blood:air and tissue:air partition coefficients using theoretical descriptors.

Human blood:air, human and rat tissue (fat, brain, liver, muscle, and kidney):air partition coefficients of a diverse set of organic compounds were correlated and predicted using structural descriptors by employing CODESSA-PRO and ISIDA programs. Four and five descriptor regression models developed using CODESSA-PRO were validated on three different test sets. Overall, these models have reasonable values of correlation coefficients (R(2)) and leave-one-out correlation coefficients (R(cv)(2)): R(2) = 0.881-0.983; R(cv)(2) = 0.826-0.962. Calculations with ISIDA resulted in models based on atom/bond sequences involving two to three atoms with statistical parameters that were similar to those of models obtained with CODESSA-PRO (R(2) = 0.911-0.974; R(cv)(2) = 0.831-0.936). A mixed pool of molecular and fragment descriptors did not lead to significant improvement of the models.

QSAR studies on 1-phenylbenzimidazoles as inhibitors of the platelet-derived growth factor.

This work is devoted to the development of quantitative structure-activity relationship (QSAR) models of the biological activity of 123 1-phenylbenzimidazoles as inhibitors of the PDGF receptor. The molecular features are represented by chemical descriptors that have been calculated on geometrical, topological, quantum mechanical, and electronic basis by using CODESSA PRO. The obtained models, linear (multilinear regression) and nonlinear (artificial neural network), are aimed to link the structures to their reported activity log 1/IC50. The former model can be used for physico-chemical interpretation, while the latter possesses a superior predictive ability.

A general treatment of solubility. 3. Principal component analysis (PCA) of the solubilities of diverse solutes in diverse solvents.

A phenomenological study of solubility has been conducted using a combination of quantitative structure-property relationship (QSPR) and principal component analysis (PCA). A solubility database of 4540 experimental data points was used that utilized available experimental data into a matrix of 154 solvents times 397 solutes. Methodology in which QSPR and PCA are combined was developed to predict the missing values and to fill the data matrix. PCA on the resulting filled matrix, where solutes are observations and solvents are variables, shows 92.55% of coverage with three principal components. The corresponding transposed matrix, in which solvents are observations and solutes are variables, showed 62.96% of coverage with four principal components.

QSAR treatment of drugs transfer into human breast milk.

A satisfactory model is developed using CODESSA PRO for the correlation and prediction of milk to plasma concentration ratios (M/P ratio) for diverse pharmaceuticals. A set of experimentally derived M/P ratio values were collected from the literature for 115 widely used pharmaceuticals. The experimental logarithmic M/P ratios were tested with more than 850 theoretical molecular descriptors including constitutional, topological, geometrical, quantum chemical, thermodynamic, and electrostatic types. Based on the data set, for 100 commonly used drugs, a seven-parameter QSAR model was derived that shows a satisfactory (R(2)=0.791) correlation between predicted and observed values of log(M/P) ratio.

The classification of solvents by combining classical QSPR methodology with principal component analysis.

The results of a quantitative structure-property relationship (QSPR) analysis of 127 different solvent scales and 774 solvents using the CODESSA PRO program are presented. QSPR models for each scale were constructed using only theoretical descriptors. The high quality of the models is reflected by the squared multiple correlation coefficients that range from 0.726 to 0.999; only 18 models have R2< 0.800. This enables direct theoretical calculation of predicted values for any scale and/or for any organic solvent, including those previously unmeasured. The molecular descriptors involved in the models are classified and discussed according to (i) the origin of their calculation (i.e., constitutional, geometric, charge-related, etc.) and (ii) the commonly accepted classification of physical interactions between the solute and solvent molecules in liquid (condensed) media. A reduced matrix 774 (solvents) x 100 (solvent scales) was selected for the principal component analysis (PCA) by taking into account only the solvent scales with more than 20 experimental data points. The first 5 principal components account for 75% of the total variance. The robustness of the PCA model obtained was validated by the comparison models development for restricted submatrices of data and with the results obtained for the full data set. The total variance accounted for by the first three PCs, for the submatrices with the same number of solvent scales but different numbers of solvents, varies from 68.2% to 59.0%. This demonstrates that the total variance described by the first 3 components is essentially stable as the number of solvents involved varies from 100 to 774. Subsequently, a matrix with 703 diverse solvents and 100 solvent scales was selected for the general classification of the solvents and scales according to the scores and loadings obtained from the PCA treatment. Classification of the theoretical molecular descriptors, derived from the chemical structure alone, according to their relevance to specific types of intermolecular interaction (cavity formation, electrostatic polarization, dispersion, and hydrogen bonding) in liquid media enables a more easily comprehensible physical interpretation of the QSPR of molecular properties in liquids and solutions. The reported QSPR models for solvent scales with theoretical molecular descriptors and the results of the PCA analysis are potentially of great practical importance, as they extend the applicability of correlations with empirical solvent scales to many previously unmeasured systems.

QSPR treatment of rat blood:air, saline:air and olive oil:air partition coefficients using theoretical molecular descriptors.

A QSPR treatment has been applied to a data set that consists of 100 diverse organic compounds to relate the logarithmic function of rat blood:air, saline:air and olive oil:air partition coefficients (denoted by log K(b:a), log K(s:a), and log K(o:a), respectively), with theoretical molecular and fragment descriptors. Three QSPR models with squared correlation coefficients of 0.881, 0.926, and 0.922, respectively, were obtained. The verification of the predictive power of these models on a test set of 33 organic chemicals that were not included in the training set gave satisfactory squared correlation coefficients: 0.791 for rat blood:air, 0.794 for saline:air and 0.846 for olive oil:air.

QSPR of 3-aryloxazolidin-2-one antibacterials.

A QSPR treatment has been applied to a data set consisting of 60 3-aryloxazolidin-2-one antibacterials to relate the in vitro minimum inhibitory concentration (MIC) (required to inhibiting growth of S. aureus) with theoretical molecular and fragment descriptors. The treatment using codessa pro descriptors leads to a seven-parameter model with r2 = 0.820 and r2cv = 0.758.

Quantitative structure-property relationship modeling of beta-cyclodextrin complexation free energies.

CODESSA-PRO was used to model binding energies for 1:1 complexation systems between 218 organic guest molecules and beta-cyclodextrin, using a seven-parameter equation with R2 = 0.796 and Rcv2 = 0.779. Fragment-based TRAIL calculations gave a better fit with R2 = 0.943 and Rcv2 = 0.848 for 195 data points in the database. The advantages and disadvantages of each approach are discussed, and it is concluded that a combination of the two approaches has much promise from a practical viewpoint.

Aqueous biphasic systems. Partitioning of organic molecules: a QSPR treatment.

The partitioning of 29 small organic probes in a PEG-2000/(NH4)2SO4 biphasic system was investigated using a quantitative structure-property relationship (QSPR) approach. A three-descriptor equation with the squared correlation coefficient (R2) of 0.97 for the partition coefficient (log D) was obtained. All descriptors were derived solely from the chemical structure of the compounds. Using the same descriptors, a three-parameter model was also obtained for log P (octanol/water, R2=0.89); predicted log P values were used as an external descriptor for modeling log D.

The present utility and future potential for medicinal chemistry of QSAR/QSPR with whole molecule descriptors.

Whole-molecule descriptors are obtained computationally from molecular structures using a variety of programs. Their applications are reviewed in the areas of solubility, bioavailability, bio- and nonbio-degradability and toxicity.