RESUMEN

Generative machine learning models have become widely adopted in drug discovery and other fields to produce new molecules and explore molecular space, with the goal of discovering novel compounds with optimized properties. These generative models are frequently combined with transfer learning or scoring of the physicochemical properties to steer generative design, yet often, they are not capable of addressing a wide variety of potential problems, as well as converge into similar molecular space when combined with a scoring function for the desired properties. In addition, these generated compounds may not be synthetically feasible, reducing their capabilities and limiting their usefulness in real-world scenarios. Here, we introduce a suite of automated tools called MegaSyn representing three components: a new hill-climb algorithm, which makes use of SMILES-based recurrent neural network (RNN) generative models, analog generation software, and retrosynthetic analysis coupled with fragment analysis to score molecules for their synthetic feasibility. We show that by deconstructing the targeted molecules and focusing on substructures, combined with an ensemble of generative models, MegaSyn generally performs well for the specific tasks of generating new scaffolds as well as targeted analogs, which are likely synthesizable and druglike. We now describe the development, benchmarking, and testing of this suite of tools and propose how they might be used to optimize molecules or prioritize promising lead compounds using these RNN examples provided by multiple test case examples.

RESUMEN

We report the development and implementation of a cheminformatics tool which aids in the design of compounds during exploratory chemistry and lead optimization. The Heterocyclic Regioisomer Enumeration and MDDR Search (HREMS) tool allows medicinal chemists to build greater structural diversity into their synthetic planning by enabling a systematic, automated enumeration of heterocyclic regioisomers of target structures. To help chemists overcome biases arising from past experience or synthetic accessibility, the HREMS tool further provides statistics on clinical testing for each enumerated regioisomer substructure using an automated search of a commercial database. Ready access to this type of information can help chemists make informed choices on the targets they will pursue being mindful of past experience with these structures in drug development. This tool and its components can be incorporated into other cheminformatics workflows to leverage their capabilities in triaging and in silico compound enumeration.

Asunto(s)

RESUMEN

5-Chloro-1,3-dihydroxyacridone, 1, is a potent and selective inhibitor of Herpes Simplex Virus Type-1 (HSV-1).(1) Substituted 1,3-dihydroxyacridones represent a new class of nonnucleoside HSV-1 inhibitors, and biochemical studies indicate a novel mechanism of action for 1, although the target is not yet defined.(2) With the goal of lead optimization, analogues of 1 were synthesized in an effort to describe the structure-activity relationships between 1 and its hypothetical binding site. Modifications of key functional groups led to the identity of several features of 1 that were important for activity. In the process, a more expedient and reliable synthesis of 1 and its analogues was developed. Analogues were evaluated against HSV-1 and HSV-2 using a viral plaque-elimination assay for viral inhibition of HSV-1 and HSV-2, and effects on replication of the host cell were also measured in order to assess a therapeutic index (TI) of selectivity. Several new analogues with significant antiviral activity were identified, including 5-methoxy-1,3-dihydroxyacridone (11), which inhibits replication of several HSV-2 strains with a mean ED(50) of 0.7 muM and a TI range of 25-60-fold.

Asunto(s)

RESUMEN

A series of analogues of the protein kinase C (PKC) inhibitory natural product balanol which bear modified benzophenone subunits are described. The analogues were designed with the goal of uncovering structure-activity features that could be used in the development of PKC inhibitors with a reduced polar character compared to balanol itself. The results of these studies suggest that most of the benzophenone features found in the natural product are important for obtaining potent PKC inhibitory compounds. However, several modifications were found to lead to selective inhibitors of the related enzyme protein kinase A (PKA), and several specific modifications to the polar structural elements of the benzophenone were found to provide potent PKC inhibitors. In particular, it was found that replacement of the benzophenone carboxylate with bioisosteric equivalents could lead to potent analogues. Further, a tolerance for lipophilic substituents on the terminal benzophenone ring was uncovered. These results are discussed in light of recently available structural information for PKA.

Asunto(s)