The recently developed efficient protocol combining implicit and explicit, accurate quantum-mechanical modeling of the condensed state (Katsyuba et al., J. Chem. Phys. 155, 024507 [2021]) is used to describe the IR spectra of liquid ethanol and its solutions in CCl4 . The relative abundance of the anti and gauche conformers of ethanol is shown to increase from ~40:60 in the gas phase to ~55:45 in the liquid phase. In spite of a moderate impact of media effects on the conformational composition of the liquid, the solvent strongly influences vibrational frequencies, IR intensities, and normal modes of each conformer, producing qualitatively different spectra compared to the gas phase and CCl4 solution. Further, these solvent effects affecting IR frequencies and intensities depend not only on the conformation of the solvated molecule but also on the solvating species. Nevertheless, vibrational frequencies of anti and gauche conformers of liquid ethanol and its several isotopomers practically coincide with each other. Convenient liquid-state conformational markers in the fingerprint region of IR spectra are revealed for the hydroxyl-deuterated species: CH3 CH2 OD, CH3 CHDOD, CH3 CD2 OD, and CD3 CD2 OD.

Buchwald-Hartwig amination reaction catalyzed by palladium plays an important role in drug synthesis. In the last few years, machine learning-assisted strategies emerged and quickly gained attention. In this article, an importance and relevance-based integrated feature screening method is proposed to effectively filter high-dimensional feature descriptor data. Then, a regularized machine learning boosting tree model, eXtreme Gradient Boosting, is introduced to intelligently predict reaction performance in multidimensional chemistry space. Furthermore, convergence, interpretability, generalization, and the internal association between reaction conditions and yields are excavated, which provides intelligent assistance for the optimal design of coupling reaction system and evaluating the reaction conditions. Compared with recently published results, the proposed method requires fewer feature descriptors, takes less time, and achieves more accurate prediction accuracy.

Identifying potential associations between proteins and compounds is significant and challenging in the drug discovery process. Existing deep-learning-based methods tend to treat compounds and proteins as sequences or graphs. Inspired by the rapid development of computer vision technology, we argue that more abundant characterizations can be extracted from the images of compounds than from their sequences or graphs. Therefore, we propose an interaction model named picture-word order compound protein interaction (PWO-CPI) which learns the representation from structural images of compounds and protein sequences by using convolutional neural network (CNN). The experiments show that PWO-CPI outperforms state-of-the-art CPI prediction models. We also perform drug-drug interaction (DDI) experiments to validate the strong potential of structural formula images of molecular structures as molecular features. In addition, with the aid of generative adversarial networks, the visualization of image features demonstrates PWO-CPI can learn compound structural features implicitly and automatically.

High-dimensional potential energy surface (PES) for van der Waals systems with spectroscopic accuracy, is of great importance for quantum dynamics and an extremely challenge job. CO-N2 is a typical van der Waals system and its high-precision PES may help elucidate weak interaction mechanisms. Taking CO-N2 potential energies calculated by CCSD(T)-F12b/aug-cc-pVQZ as the benchmark, we establish an accurate, robust, and efficient machine learning model by using only four molecular structure descriptors based on 7966 benchmark potential energies. The highest accuracy is obtained by a stacking ensemble DNN (SeDNN). Its evaluation parameters MAE, RMSE, and R2 reach 0.096, 0.163, 0.9999 cm-1 , respectively, and the spectroscopic accuracy for vibration spectrum is achieved with predicted PES, which shows SeDNN superior goodness-of-fit and prediction performance. An elaborated PES with the reported global minimum has been predicted with the model, which perfectly reproduces CCSD(T) potential energies and the analytical MLR PES [PCCP, 2018, 20, 2036]. The critical points (global minimum, TSI, TSII, and their barriers), potential curve, and entire PES profile are remarkably consistent with CCSD(T) calculations. To further improve the usability of constructing PESs in practice, the size of the training set (energy points) for the model is reduced to 50%, 30%, and 20% of the database, respectively. The results show that even training with the smallest training set (1593 points), the PES only differs 2.555 cm-1 with the analytic MLR PES. Therefore, the proposed SeDNN is promisingly an alternative efficient tool to construct subtle PES for van der Waals systems.

Traditionally, substituents are described not by their intrinsic properties, but by their effect elsewhere in a molecule. However, the quantum theory of atoms in molecules (QTAIM) provides a route to intrinsic substituent descriptors. Ideally, these descriptors would exhibit minimal change as the local environment changes, and hence be considered transferable. Whether this is true is an open question. Here, we evaluated the transferability of QTAIM functional group descriptors for 117 functional groups in a series of 17 substrates to determine whether descriptors obtained using hydrogen as substrate are transferable. The functional group volume has a strong, consistent, linear relationship throughout. All other hydrogen-based group descriptors exhibit a relatively strong linear relationship with those in carbon-based substrates and a reasonable linear relationship with those in non-carbon-based substrates. Outliers are readily interpretable in terms of substrate induced functional group geometry changes. As expected, directional descriptors lying along the substituent-substrate axis are the least conserved.

The coinage-metal clusters possess a natural complexity in their theoretical treatment that may be accompanied by inherent shortcomings in the methodological approach. Herein, we performed a scalar-relativistic density functional theory study, considering Perdew, Burke, and Ernzerhof (PBE) with (empirical and semi empirical) van der Waals (vdW), spin-orbit coupling (SOC), +U (Hubbard term), and their combinations, to treat the Cu13 , Ag13 , and Au13 clusters in different structural motifs. The energetic scenario is given by the confirmation of the 3D lowest energy configurations for Cu13 and Ag13 within all approaches, while for Au13 there is a 2D/3D competition, depending on the applied correction. The 2D geometry is 0.43 eV more stable with plain PBE than the 3D one, the SOC, +U, and/or vdW inclusion decreases the overestimated stability of the planar configurations, where the most surprising result is found by the D3 and D3BJ vdW corrections, for which the 3D configuration is 0.29 and 0.11 eV, respectively, more stable than the 2D geometry (with even higher values when SOC and/or +U are added). The D3 dispersion correction represents 7.9% (4.4%) of the total binding energy for the 3D (2D) configuration, (not) being enough to change the sd hybridization and the position of the occupied d -states. Our predictions are in agreement with experimental results and in line with the best results obtained for bulk systems, as well as with hybrid functionals within D3 corrections. The properties description undergoes small corrections with the different approaches, where general trends are maintained, that is, the average bond length is smaller (larger) for lower (higher)-coordinated structures, since a same number of electrons are shared by a smaller (larger) number of bonds, consequently, the bonds are stronger (weaker) and shorter (longer) and the sd hybridization index is larger (smaller). Thus, Au has a distinct behavior in relation to its lighter congeners, with a complex potential energy surface, where in addition to the relevant relativistic effects, correlation and dispersion effects must also be considered.

Due to the role of loops in protein function, loop modeling is an important problem in computational biology. We present a new approach to loop modeling based on a combinatorial version of distance geometry, where the search space of the associated problem is represented by a binary tree and a branch-and-prune method is defined to explore it, following an atomic ordering previously given. This ordering is used to calculate the coordinates of atoms from the positions of its predecessors. In addition to the theoretical development, computational results are presented to illustrate the advantage of the proposed method, compared with another approach of the literature. Our algorithm is freely available at https://github.com/michaelsouza/bpl.

There has been tremendous advancement in machine learning (ML) applications in computational chemistry, particularly in neural network potentials (NNP). NNPs can approximate potential energy surface (PES) as a high dimensional function by learning from existing reference data, thereby circumventing the need to solve the electronic Schrödinger equation explicitly. As a result, ML accelerates chemical space exploration and property prediction compared to quantum mechanical methods. Novel ML methods have the potential to provide efficient means for predicting the properties of molecules. However, this potential has been limited by the lack of standard comparative evaluations. In this work, we compare four selected models, that is, ANI, PhysNet, SchNet, and BAND-NN, developed to represent the PES of small organic molecules. We evaluate these models for their accuracy and transferability on two different test sets (i) Small organic molecules of up to eight-heavy atoms on which ANI and SchNet achieve root mean square error (RMSE) of 0.55 and 0.60 kcal/mol, respectively. (ii) On random selection of molecules from the GDB-11 database with 10-heavy atoms, ANI achieves RMSE of 1.17 kcal/mol and SchNet achieves RMSE of 1.89 kcal/mol. We examine their ability to produce smooth meaningful surface by performing PES scans for bond stretch, angle bend, and dihedral rotations on relatively large molecules to assess their possible application in molecular dynamics simulations. We also evaluate their performance for yielding minimum energy structures via geometry optimization using various minimization algorithms. All these models were also able to accurately differentiate different isomers of the same empirical formula C10H20 . ANI and PhysNet achieve an RMSE of 0.29 and 0.52 kcal/mol, respectively, on C10H20 isomers.

Since phospholipids are the most important components in the structure of biomembranes, they deserve to be considered with a lot of attention in both experimental and computational theoretical studies using molecular simulation methods related to the research in the fields of drug design and drug delivery where they involve knowledge about the interactions of drug molecules with cell membranes. To employ the molecular simulation approach for this purpose the essential requirement is having information about the initial structure of phospholipids and how they interact with the drugs. Therefore in this article, we introduce an open-source software package in Python programming language for utilizing data manipulation for generation and developing the initial structure of biomolecular cells to provide the needed information for investigation in drug delivery systems. In addition, the proposed software package can be used for the efficient storage of membrane structural data to be exploited in designing new drug delivery systems. To verify the performance of the code and the results of the simulations, several analyses have been done, such as the calculation of area per lipid and self-diffusion coefficient, in addition to lipid order parameter. The results were in complete agreement with the references.

Explicit treatment of electronic polarizability in empirical force fields (FFs) represents an extension over a traditional additive or pairwise FF and provides a more realistic model of the variations in electronic structure in condensed phase, macromolecular simulations. To facilitate utilization of the polarizable FF based on the classical Drude oscillator model, Drude Prepper has been developed in CHARMM-GUI. Drude Prepper ingests additive CHARMM protein structures file (PSF) and pre-equilibrated coordinates in CHARMM, PDB, or NAMD format, from which the molecular components of the system are identified. These include all residues and patches connecting those residues along with water, ions, and other solute molecules. This information is then used to construct the Drude FF-based PSF using molecular generation capabilities in CHARMM, followed by minimization and equilibration. In addition, inputs are generated for molecular dynamics (MD) simulations using CHARMM, GROMACS, NAMD, and OpenMM. Validation of the Drude Prepper protocol and inputs is performed through conversion and MD simulations of various heterogeneous systems that include proteins, nucleic acids, lipids, polysaccharides, and atomic ions using the aforementioned simulation packages. Stable simulations are obtained in all studied systems, including 5 μs simulation of ubiquitin, verifying the integrity of the generated Drude PSFs. In addition, the ability of the Drude FF to model variations in electronic structure is shown through dipole moment analysis in selected systems. The capabilities and availability of Drude Prepper in CHARMM-GUI is anticipated to greatly facilitate the application of the Drude FF to a range of condensed phase, macromolecular systems.

Linear molecules such as CO2 contain a positive π-hole ring that surrounds C on the molecule's equator. Quantum calculations examine the question as to how many bases can simultaneously bind to this ring. Linear molecules examined are TO2 , where T = C, Si, Ge, Sn; bases are NCH and NH3 . CO2 engages in the weakest of the tetrel bonds, and can bind up to three NCH and two NH3 . Unlike σ-hole tetrel bonds, Si forms the strongest tetrel bonds, with interaction energies as high as 43 kcal/mol with NH3 . But like GeO2 , SiO2 can sustain only two bases in its equatorial ring. The π-hole ring of SnO2 can engage in up to four tetrel bonds with either NCH or NH3 , even though these bonds are weaker than those with GeO2 or SiO2 . As all of these complexes cast TO2 in the role of multiple electron acceptor, the resulting negative cooperativity makes each successive bond weaker than its predecessor as bases are added, as well as reducing the magnitude of the central molecule's π-hole.

Composites of carbon nanotubes (CNTs) and lignin are promising and potentially cheap precursors of-to this day-expensive carbon fibers. Since the control of the CNT-lignin interface is crucial to maximize fiber performance, it is imperative to understand the fundamental noncovalent interactions between lignin and CNT. In the present study a density functional theory study is conducted to investigate the fundamental noncovalent interaction strength between metallic (n, n) single-walled CNT (SWCNT) and simple lignin model molecules. In particular, the respective adsorption energies are used to gauge the strength of interaction classes (ππ interaction, CHπ hydrogen bonding and OH-related hydrogen bonding. From the data, substituent-dependent interaction trends as well as class- and curvature-dependent interaction trends are derived. Overall, we find that most of the interaction strength trends appear to be strongly influenced by geometry: flat orientation of the test molecules relative to the (n, n) SWCNT surface and small (n, n) SWCNT curvature-that is, large diameter enhances the CHπ and ππ interactions.