This work introduces a novel application of generative adversarial networks (GANs) for the prediction of starting geometries in transition state (TS) searches based on the geometries of reactants and products. The multi-dimensional potential energy space of a chemical reaction often complicates the location of a starting TS geometry, leading to the correct TS combining reactants and products in question. The proposed TS-GAN efficiently maps the space between reactants and products and generates reliable TS guess geometries, and it can be easily combined with any quantum chemical software package performing geometry optimizations. The TS-GAN was trained and applied to generate TS guess structures for typical chemical reactions, such as hydrogen migration, isomerization, and transition metal-catalyzed reactions. The performance of the TS-GAN was directly compared to that of classical approaches, proving its high accuracy and efficiency. The current TS-GAN can be extended to any dataset that contains sufficient chemical reactions for training. The software is freely available for training, experimentation, and prediction at https://github.com/ekraka/TS-GAN.

Computational prediction of Protein-Ligand Interaction (PLI) is an important step in the modern drug discovery pipeline as it mitigates the cost, time, and resources required to screen novel therapeutics. Deep Neural Networks (DNN) have recently shown excellent performance in PLI prediction. However, the performance is highly dependent on protein and ligand features utilized for the DNN model. Moreover, in current models, the deciphering of how protein features determine the underlying principles that govern PLI is not trivial. In this work, we developed a DNN framework named SSnet that utilizes secondary structure information of proteins extracted as the curvature and torsion of the protein backbone to predict PLI. We demonstrate the performance of SSnet by comparing against a variety of currently popular machine and non-Machine Learning (ML) models using various metrics. We visualize the intermediate layers of SSnet to show a potential latent space for proteins, in particular to extract structural elements in a protein that the model finds influential for ligand binding, which is one of the key features of SSnet. We observed in our study that SSnet learns information about locations in a protein where a ligand can bind, including binding sites, allosteric sites and cryptic sites, regardless of the conformation used. We further observed that SSnet is not biased to any specific molecular interaction and extracts the protein fold information critical for PLI prediction. Our work forms an important gateway to the general exploration of secondary structure-based Deep Learning (DL), which is not just confined to protein-ligand interactions, and as such will have a large impact on protein research, while being readily accessible for de novo drug designers as a standalone package.

Over the past two decades, the vibrational Stark effect has become an important tool to measure and analyze the in situ electric field strength in various chemical environments with infrared spectroscopy. The underlying assumption of this effect is that the normal stretching mode of a target bond such as CO or CN of a reporter molecule (termed vibrational Stark effect probe) is localized and free from mass-coupling from other internal coordinates, so that its frequency shift directly reflects the influence of the vicinal electric field. However, the validity of this essential assumption has never been assessed. Given the fact that normal modes are generally delocalized because of mass-coupling, this analysis was overdue. Therefore, we carried out a comprehensive evaluation of 68 vibrational Stark effect probes and candidates to quantify the degree to which their target normal vibration of probe bond stretching is decoupled from local vibrations driven by other internal coordinates. The unique tool we used is the local mode analysis originally introduced by Konkoli and Cremer, in particular the decomposition of normal modes into local mode contributions. Based on our results, we recommend 31 polyatomic molecules with localized target bonds as ideal vibrational Stark effect probe candidates.

The intrinsic bonding nature of ${\lambda }^3$ -iodanes was investigated to determine where its hypervalent bonds fit along the spectrum between halogen bonding and covalent bonding. Density functional theory with an augmented Dunning valence triple zeta basis set ( $\omega$ B97X-D/aug-cc-pVTZ) coupled with vibrational spectroscopy was utilized to study a diverse set of 34 hypervalent iodine compounds. This level of theory was rationalized by comparing computational and experimental data for a small set of closely-related and well-studied iodine molecules and by a comparison with CCSD(T)/aug-cc-pVTZ results for a subset of the investigated iodine compounds. Axial bonds in ${\lambda }^3$ -iodanes fit between the three-center four-electron bond, as observed for the trihalide species IF ${}_2^{-}$ and the covalent FI molecule. The equatorial bonds in ${\lambda }^3$ -iodanes are of a covalent nature. We explored how the equatorial ligand and axial substituents affect the chemical properties of ${\lambda }^3$ -iodanes by analyzing natural bond orbital charges, local vibrational modes, the covalent/electrostatic character, and the three-center four-electron bonding character. In summary, our results show for the first time that there is a smooth transition between halogen bonding → 3c–4e bonding in trihalides → 3c–4e bonding in hypervalent iodine compounds → covalent bonding, opening a manifold of new avenues for the design of hypervalent iodine compounds with specific properties.

Severe Acute Respiratory Syndrome Corona Virus 2 has altered life on a global scale. A concerted effort from research labs around the world resulted in the identification of potential pharmaceutical treatments for CoVID-19 using existing drugs, as well as the discovery of multiple vaccines. During an urgent crisis, rapidly identifying potential new treatments requires global and cross-discipline cooperation, together with an enhanced open-access research model to distribute new ideas and leads. Herein, we introduce an application of a deep neural network based drug screening method, validating it using a docking algorithm on approved drugs for drug repurposing efforts, and extending the screen to a large library of 750,000 compounds for de novo drug discovery effort. The results of large library screens are incorporated into an open-access web interface to allow researchers from diverse fields to target molecules of interest. Our combined approach allows for both the identification of existing drugs that may be able to be repurposed and de novo design of ACE2-regulatory compounds. Through these efforts we demonstrate the utility of a new machine learning algorithm for drug discovery, SSnet, that can function as a tool to triage large molecular libraries to identify classes of molecules with possible efficacy.