The AI-driven Drug Design (AIDD) platform: an interactive multi-parameter optimization system integrating molecular evolution with physiologically based pharmacokinetic simulations.

Computer-aided drug design has advanced rapidly in recent years, and multiple instances of in silico designed molecules advancing to the clinic have demonstrated the contribution of this field to medicine. Properly designed and implemented platforms can drastically reduce drug development timelines and costs. While such efforts were initially focused primarily on target affinity/activity, it is now appreciated that other parameters are equally important in the successful development of a drug and its progression to the clinic, including pharmacokinetic properties as well as absorption, distribution, metabolic, excretion and toxicological (ADMET) properties. In the last decade, several programs have been developed that incorporate these properties into the drug design and optimization process and to varying degrees, allowing for multi-parameter optimization. Here, we introduce the Artificial Intelligence-driven Drug Design (AIDD) platform, which automates the drug design process by integrating high-throughput physiologically-based pharmacokinetic simulations (powered by GastroPlus) and ADMET predictions (powered by ADMET Predictor) with an advanced evolutionary algorithm that is quite different than current generative models. AIDD uses these and other estimates in iteratively performing multi-objective optimizations to produce novel molecules that are active and lead-like. Here we describe the AIDD workflow and details of the methodologies involved therein. We use a dataset of triazolopyrimidine inhibitors of the dihydroorotate dehydrogenase from Plasmodium falciparum to illustrate how AIDD generates novel sets of molecules.

Design and tests of prospective property predictions for novel antimalarial 2-aminopropylaminoquinolones.

There is a pressing need to improve the efficiency of drug development, and nowhere is that need more clear than in the case of neglected diseases like malaria. The peculiarities of pyrimidine metabolism in Plasmodium species make inhibition of dihydroorotate dehydrogenase (DHODH) an attractive target for antimalarial drug design. By applying a pair of complementary quantitative structure-activity relationships derived for inhibition of a truncated, soluble form of the enzyme from Plasmodium falciparum (s-PfDHODH) to data from a large-scale phenotypic screen against cultured parasites, we were able to identify a class of antimalarial leads that inhibit the enzyme and abolish parasite growth in blood culture. Novel analogs extending that class were designed and synthesized with a goal of improving potency as well as the general pharmacokinetic and toxicological profiles. Their synthesis also represented an opportunity to prospectively validate our in silico property predictions. The seven analogs synthesized exhibited physicochemical properties in good agreement with prediction, and five of them were more active against P. falciparum growing in blood culture than any of the compounds in the published lead series. The particular analogs prepared did not inhibit s-PfDHODH in vitro, but advanced biological assays indicated that other examples from the class did inhibit intact PfDHODH bound to the mitochondrial membrane. The new analogs, however, killed the parasites by acting through some other, unidentified mechanism 24-48 h before PfDHODH inhibition would be expected to do so.

Modeling ADMET.

Drug discovery and development is a costly and time-consuming endeavor (Calcoen et al. Nat Rev Drug Discov 14(3):161-162, 2015; The truly staggering cost of inventing new drugs. Forbes. http://www.forbes.com/sites/matthewherper/2012/02/10/the-truly-staggering-cost-of-inventing-new-drugs/, 2012; Scannell et al. Nat Rev Drug Discov 11(3):191-200, 2012). Over the last two decades, computational tools and in silico models to predict ADMET (Adsorption, Distribution, Metabolism, Excretion, and Toxicity) profiles of molecules have been incorporated into the drug discovery process mainly in an effort to avoid late-stage failures due to poor pharmacokinetics and toxicity. It is now widely recognized that ADMET issues should be addressed as early as possible in drug discovery. Here, we describe in detail how ADMET models can be developed and applied using a commercially available package, ADMET Predictor™ 7.2 (ADMET Predictor v7.2. Simulations Plus, Inc., Lancaster, CA, USA).

Using Cheminformatics in Drug Discovery.

This chapter illustrates how cheminformatics can be applied to designing novel compounds that are active at the primary target and have good predicted ADMET properties. Examples of various cheminformatics techniques are illustrated in the process of designing inhibitors that inhibit both cyclooxygenase isoforms but are more potent toward COX-2. The first step in the process is to create a knowledge database of cyclooxygenase inhibitors in the public domain. This data was analyzed to find activity cliffs - small structural changes that result in drastic changes in potency. Additional cyclooxygenase potency and selectivity trends were obtained using matched molecular pair analysis. QSAR models were then developed to predict cyclooxygenase potency and selectivity. Next, computational algorithms were used to generate novel scaffolds starting from known cyclooxygenase inhibitors. Nine virtual libraries containing 240 compounds each were constructed. Predictions from the cyclooxygenase QSAR models were used to eliminate molecules with undesirable potency or selectivity. Additionally, the compounds were screened in silico for undesirable ADMET properties, e.g., low solubility, permeability, metabolic stability, or high toxicity, using a liability scoring system known as ADMET Risk™. Eight synthetic candidates were identified from this process after incorporating knowledge gained from activity cliff analysis. Four of the compounds were synthesized and tested to measure their COX-1 and COX-2 IC(50) values as well as several ADME properties. The best compound, SLP0020, had a COX-1 IC(50) of 770 nM and COX-2 IC(50) of 130 nM.

Using beta binomials to estimate classification uncertainty for ensemble models.

BACKGROUND: Quantitative structure-activity (QSAR) models have enormous potential for reducing drug discovery and development costs as well as the need for animal testing. Great strides have been made in estimating their overall reliability, but to fully realize that potential, researchers and regulators need to know how confident they can be in individual predictions. RESULTS: Submodels in an ensemble model which have been trained on different subsets of a shared training pool represent multiple samples of the model space, and the degree of agreement among them contains information on the reliability of ensemble predictions. For artificial neural network ensembles (ANNEs) using two different methods for determining ensemble classification - one using vote tallies and the other averaging individual network outputs - we have found that the distribution of predictions across positive vote tallies can be reasonably well-modeled as a beta binomial distribution, as can the distribution of errors. Together, these two distributions can be used to estimate the probability that a given predictive classification will be in error. Large data sets comprised of logP, Ames mutagenicity, and CYP2D6 inhibition data are used to illustrate and validate the method. The distributions of predictions and errors for the training pool accurately predicted the distribution of predictions and errors for large external validation sets, even when the number of positive and negative examples in the training pool were not balanced. Moreover, the likelihood of a given compound being prospectively misclassified as a function of the degree of consensus between networks in the ensemble could in most cases be estimated accurately from the fitted beta binomial distributions for the training pool. CONCLUSIONS: Confidence in an individual predictive classification by an ensemble model can be accurately assessed by examining the distributions of predictions and errors as a function of the degree of agreement among the constituent submodels. Further, ensemble uncertainty estimation can often be improved by adjusting the voting or classification threshold based on the parameters of the error distribution. Finally, the profiles for models whose predictive uncertainty estimates are not reliable provide clues to that effect without the need for comparison to an external test set.