Explaining Blood-Brain Barrier Permeability of Small Molecules by Integrated Analysis of Different Transport Mechanisms.

The blood-brain barrier (BBB) represents a major obstacle to delivering drugs to the central nervous system (CNS), resulting in the lack of effective treatment for many CNS diseases including brain cancer. To accelerate CNS drug development, computational prediction models could save the time and effort needed for experimental evaluation. Here, we studied BBB permeability focusing on active transport (influx and efflux) as well as passive diffusion using previously published and self-curated data sets. We created prediction models based on physicochemical properties, molecular substructures, or their combination to understand which mechanisms contribute to BBB permeability. Our results show that features that predicted passive diffusion over membranes overlap with features that explain endothelial permeation of approved CNS-active drugs. We also identified physical properties and molecular substructures that positively or negatively predicted BBB transport. These findings provide guidance toward identifying BBB-permeable compounds by optimally matching physicochemical and molecular properties to BBB transport mechanisms.

Reprogramming Nonribosomal Peptide Synthetases for Site-Specific Insertion of α-Hydroxy Acids.

High-throughput engineering has the potential to revolutionize the customization of biosynthetic assembly lines for the sustainable production of pharmaceutically relevant natural product analogs. Here, we show that the substrate specificity of gatekeeper adenylation domains of nonribosomal peptide synthetases can be switched from an α-amino acid to an α-hydroxy acid in a single round of combinatorial mutagenesis and selection using yeast cell surface display. In addition to shedding light on how such proteins discriminate between amino and hydroxy groups, the remodeled domains function in a pathway context to produce α-hydroxy acid-containing linear peptides and cyclic depsipeptides with high efficiency. Site-specific replacement of backbone amines with oxygens by an engineered synthetase provides the means to probe and tune the activities of diverse peptide metabolites in a simple and predictable fashion.

PaccMannRL: De novo generation of hit-like anticancer molecules from transcriptomic data via reinforcement learning.

With the advent of deep generative models in computational chemistry, in-silico drug design is undergoing an unprecedented transformation. Although deep learning approaches have shown potential in generating compounds with desired chemical properties, they disregard the cellular environment of target diseases. Bridging systems biology and drug design, we present a reinforcement learning method for de novo molecular design from gene expression profiles. We construct a hybrid Variational Autoencoder that tailors molecules to target-specific transcriptomic profiles, using an anticancer drug sensitivity prediction model (PaccMann) as reward function. Without incorporating information about anticancer drugs, the molecule generation is biased toward compounds with high predicted efficacy against cell lines or cancer types. The generation can be further refined by subsidiary constraints such as toxicity. Our cancer-type-specific candidate drugs are similar to cancer drugs in drug-likeness, synthesizability, and solubility and frequently exhibit the highest structural similarity to compounds with known efficacy against these cancer types.

Prediction of Spin Density, Baird-Antiaromaticity, and Singlet-Triplet Energy Gap in Triplet-State Polybenzenoid Systems from Simple Structural Motifs.

Triplet-state aromaticity has been recently proposed as a strategy for designing functional organic electronic compounds, many of which are polycyclic aromatic systems. However, in many cases, the aromatic nature of the triplet state cannot be easily predicted. Moreover, it is often unclear how specific structural manipulations affect the electronic properties of the excited-state compounds. Herein, the relationship between the structure of polybenzenoid hydrocarbons (PBHs) and their spin-density distribution and aromatic character in the first triplet excited state is studied. Although a direct link is not immediately visible, classifying the PBHs according to their annulation sequence reveals regularities. Based on these, a set of guidelines is defined to qualitatively predict the location of spin and paratropicity and the singlet-triplet energy gap in larger PBHs, using only their smaller tri- and tetracyclic components, and subsequently tested on larger systems.