Journal of Chemical Information and Modeling 最新文献

Superphanes are a group of organic molecules from the cyclophane family. They are characterized by the presence of two parallel benzene rings joined together by six bridges. If these bridges are sufficiently long, the superphane cavity can be large enough to trap small molecules or ions. Using ab initio (time scale of 80 ps) and classical (up to 200 ns) molecular dynamics (MD) methods, we study the behavior of five fundamental molecules (M = H₂O, NH₃, HF, HCN, MeOH) encapsulated inside the experimentally reported lantern-like superphane and its two derivatives featuring slightly modified side bridges. The main focus is studying the dynamics of hydrogen bonds between the trapped M molecule and the imino nitrogen atoms of the side chains of the host superphane. The length of the N···H hydrogen bond increases in the following order: HF < HCN < H₂O < MeOH < NH₃. The mobility of the trapped molecule and its preferred position inside the superphane cage depend not only on the type of this molecule but also largely on the in/out conformational arrangement of the imino nitrogens in the side chains of the superphane. Their inward-pointing positions allow the formation of strong N···H hydrogen bonds. For this reason, these nitrogens are the preferred sites of interaction. The mobility of the molecules and their residence times on each side of the superphane have been explained by referring to the symmetry and conformation of the given superphane cage. All force field MD simulations have shown that the encapsulated molecule remained inside the superphane cage for 200 ns without any escape event to the outside. Moreover, our simulations based on some endohedral complexes in the water box also showed no exchange event. Thus, the superphanes we study are true carcerand molecules. We attribute this property to the hydrophobic side chains and their pinwheel arrangement, which makes the side walls of the studied superphanes fairly impenetrable to small molecules.

The human immunodeficiency virus presents a significant global health challenge due to its rapid mutation and the development of resistance mechanisms against antiretroviral drugs. Recent studies demonstrate the impressive performance of machine learning (ML) and deep learning (DL) models in predicting the drug resistance profile of specific FDA-approved inhibitors. However, generalizing ML and DL models to learn not only from isolates but also from inhibitor representations remains challenging for HIV-1 infection. We propose a novel drug-isolate-fold change (DIF) model framework that aims to predict drug resistance score directly from the protein sequence and inhibitor representation. Various ML and DL models, inhibitor representations, and protein representations were analyzed through realistic validation mechanisms. To enhance the molecular learning capacity of DIF models, we employ a transfer learning approach by pretraining a graph neural network (GNN) model for activity prediction on a data set of 4855 HIV-1 protease inhibitors (PIs). By performing various realistic validation strategies on internal and external genotype-phenotype data sets, we statistically show that the learned representations of inhibitors improve the predictive ability of DIF-based ML and DL models. We achieved an accuracy of 0.802, AUROC of 0.874, and r of 0.727 for the unseen external PIs. By comparing the DIF-based models with a null model consisting of isolate-fold change (IF) architecture, it is observed that the DIF models significantly benefit from molecular representations. Combined results from various testing strategies and statistical tests confirm the effectiveness of DIF models in testing novel PIs for drug resistance in the presence of an isolate.

Molecular flexibility is a commonly used, but not easily quantified term. It is at the core of understanding composition and size of a conformational ensemble and contributes to many molecular properties. For many computational workflows, it is necessary to reduce a conformational ensemble to meaningful representatives, however defining them and guaranteeing the ensemble's completeness is difficult. We introduce the concepts of torsion angular bin strings (TABS) as a discrete vector representation of a conformer's dihedral angles and the number of possible TABS (nTABS) as an estimation for the ensemble size of a molecule, respectively. Here, we show that nTABS corresponds to an upper limit for the size of the conformational space of small molecules and compare the classification of conformer ensembles by TABS with classifications by RMSD. Overcoming known drawbacks like the molecular size dependency and threshold picking of the RMSD measure, TABS is shown to meaningfully discretize the conformational space and hence allows e.g. for fast checks of the coverage of the conformational space. The current proof-of-concept implementation is based on the ETKDGv3 conformer generator as implemented in the RDKit and known torsion preferences extracted from small-molecule crystallographic data.

Under the selective pressure of nirmatrelvir, a peptidomimetic covalent drug targeting SARS-CoV-2 M^pro, various drug-resistant mutations on M^pro have been acquired in vitro. Among the mutations, L50F and E166V, along with the combination of L50F and E166V, are particularly representative and pose considerable obstacles to the effective treatment of COVID-19. Our previous study identified NMI-001 and NMI-002 as novel nonpeptide inhibitors that target SARS-CoV-2 M^pro, possessing unique scaffolds and binding modes different from those of nirmatrelvir. In view of these findings, we proposed a drug design strategy aimed at rapidly identifying inhibitors that can combat mutation-induced drug resistance. Initially, molecular dynamics (MD) simulation was employed to investigate the binding mechanisms of NMI-001 and NMI-002 against the three drug-resistant mutants (M^pro_L50F, M^pro_E166V, and M^pro_L50F+E166V). Then, we conducted two phases of high-throughput virtual screening. In the first phase, NMI-001 served as a template to perform scaffold hopping-based similarity search in a library of 15,742,661 compounds. In the second phase, 968 compounds exhibiting similarity to NMI-001 were evaluated via molecular docking and MD simulations. Six compounds that may be effective against at least one mutant were identified, and five compounds were procured for conducting in vitro assays. Finally, the compound Z1557501297 (NMI-003) exhibiting inhibitory effects against the E166V (IC₅₀ = 27.81 ± 2.65 μM) and L50F+E166V (IC₅₀ = 8.78 ± 0.74 μM) mutants was discovered. The binding modes referring to NMI-003-M^pro_E166V and NMI-003-M^pro_L50F+E166V were further elucidated at the atomic level. In summary, NMI-003 reported herein is the first compound with activity against E166V and L50F+E166V, which provides a good starting point to design novel antiviral drugs for the treatment of drug-resistant SARS-CoV-2.