Designing a molecule with desired properties is one of the biggest challenges in drug development, as it requires optimization of chemical compound structures with respect to many complex properties. To improve the compound design process, we introduce Mol-CycleGAN—a CycleGAN-based model that generates optimized compounds with high structural similarity to the original ones. Namely, given a molecule our model generates a structurally similar one with an optimized value of the considered property. We evaluate the performance of the model on selected optimization objectives related to structural properties (presence of halogen groups, number of aromatic rings) and to a physicochemical property (penalized logP). In the task of optimization of penalized logP of drug-like molecules our model significantly outperforms previous results.

Designing compounds with desired properties is a key element of the drug discovery process. However, measuring progress in the field has been challenging due to the lack of realistic retrospective benchmarks, and the large cost of prospective validation. To close this gap, we propose a benchmark based on docking, a widely used computational method for assessing molecule binding to a protein. Concretely, the goal is to generate drug-like molecules that are scored highly by SMINA, a popular docking software. We observe that various graph-based generative models fail to propose molecules with a high docking score when trained using a realistically sized training set. This suggests a limitation of the current incarnation of models for de novo drug design. Finally, we also include simpler tasks in the benchmark based on a simpler scoring function. We release the benchmark as an easy to use package available at https://github.com/cieplinski-tobiasz/smina-docking-benchmark. We hope that our benchmark will serve as a stepping stone toward the goal of automatically generating promising drug candidates.

Graph neural networks have recently become a standard method for analyzing chemical compounds. In the field of molecular property prediction, the emphasis is now on designing new model architectures, and the importance of atom featurization is oftentimes belittled. When contrasting two graph neural networks, the use of different representations possibly leads to incorrect attribution of the results solely to the network architecture. To better understand this issue, we compare multiple atom representations by evaluating them on the prediction of free energy, solubility, and metabolic stability using graph convolutional networks. We discover that the choice of atom representation has a significant impact on model performance and that the optimal subset of features is task-specific. Additional experiments involving more sophisticated architectures, including graph transformers, support these findings. Moreover, we demonstrate that some commonly used atom features, such as the number of neighbors or the number of hydrogens, can be easily predicted using only information about bonds and atom type, yet their explicit inclusion in the representation has a positive impact on model performance. Finally, we explain the predictions of the best-performing models to better understand how they utilize the available atomic features.

Interpolating between points is a problem connected simultaneously with finding geodesics and study of generative models. In the case of geodesics, we search for the curves with the shortest length, while in the case of generative models, we typically apply linear interpolation in the latent space. However, this interpolation uses implicitly the fact that Gaussian is unimodal. Thus, the problem of interpolating in the case when the latent density is non-Gaussian is an open problem. In this article, we present a general and unified approach to interpolation, which simultaneously allows us to search for geodesics and interpolating curves in latent space in the case of arbitrary density. Our results have a strong theoretical background based on the introduced quality measure of an interpolating curve. In particular, we show that maximizing the quality measure of the curve can be equivalently understood as a search of geodesic for a certain redefinition of the Riemannian metric on the space. We provide examples in three important cases. First, we show that our approach can be easily applied to finding geodesics on manifolds. Next, we focus our attention in finding interpolations in pretrained generative models. We show that our model effectively works in the case of arbitrary density. Moreover, we can interpolate in the subset of the space consisting of data possessing a given feature. The last case is focused on finding interpolation in the space of chemical compounds.