In

Reaction prediction and retrosynthesis are the cornerstones of organic chemistry. Rule-based expert systems have been most widespread approach to computationally solve these two related challenges date. However, reaction rules often fail because they ignore molecular context, which leads reactivity conflicts. Herein, we report that deep neural networks can learn resolve conflicts prioritize suitable transformation rules. We show by training our model on 3.5 million reactions taken from...

The ability to reason beyond established knowledge allows organic chemists solve synthetic problems and invent novel transformations. Herein, we propose a model that mimics chemical reasoning, formalises reaction prediction as finding missing links in graph. We have constructed graph containing 14.4 million molecules 8.2 binary reactions, which represents the bulk of all reactions ever published scientific literature. Our outperforms rule-based expert system task for 180 000 randomly...

Abstract The suitability of a hybrid density functional to qualitatively reproduce geometric and energetic details parallel π‐stacked aromatic complexes is presented. includes an ad hoc mixture half the exact (HF) exchange with uniform electron gas exchange, plus Lee, Yang, Parr's expression for correlation energy. This functional, in combination polarized, diffuse basis sets, gives binding energy parallel‐displaced benzene dimer good agreement best available high‐level calculations reported...

A data set of 19 second-row transition-metal complexes has been collated from sufficiently precise gas-phase electron-diffraction experiments and used for evaluating errors in DFT optimized geometries. Equilibrium geometries have computed using 15 different combinations exchange-correlation functionals conjunction with up to three effective core potentials. Most levels beyond the local density approximation can reproduce 29 metal−ligand bond distances selected this reasonable accuracy...

β-sheet-encoded anionic and cationic dendritic peptide amphiphiles form supramolecular copolymers when self-assembled in a 1:1 feed ratio of the monomers. These ampholytic materials have been designed for on-off polymerization response to pH triggers. The cooperative self-assembly process is switched on at physiologically relevant value can be off by increasing or decreasing value.

Multiscale methods have enjoyed a well‐celebrated place in the computational chemists’ toolbox, while next generation of so‐called adaptive quantum mechanics/molecular mechanics ( QM / MM ) were being developed on fringe for past 20 years. Adaptive hold promise extending range applicability, currently at expense complexity and scaling. There are number challenges area; firstly, ability to partition system fly has led distance, number, density, stress‐based approaches. Secondly, that smoothen...

We present a density-based adaptive quantum mechanical/molecular mechanical (DBA-QM/MM) method, whereby molecules can switch layers from the QM to MM region and vice versa. The partitioning of molecular system ensures that layer assignment change during optimization procedure, is, on fly. molecule is determined if there an absence noncovalent interactions any atom core region. presence/absence by analysis reduced density gradient. Therefore, location QM/MM boundary based physical arguments,...

In de novo drug design, computational strategies are used to generate novel molecules with good affinity the desired biological target. this work, we show that recurrent neural networks can be trained as generative models for molecular structures, similar statistical language in natural processing. We demonstrate properties of generated correlate very well train model. order enrich libraries active towards a given target, propose fine-tune model small sets molecules, which known against...

Abstract 51 V NMR chemical shifts calculated from QM/MM‐optimized (QM=quantum mechanical; MM=molecular mechanical) models of vanadium‐dependent chloroperoxidase (VCPO) are presented. An extensive number protonation states for the vanadium cofactor (active site protein) and a probable positional isomers each considered. The size QM region is increased incrementally to observe convergence behavior shifts. A total 40 assessed by comparison experimental solid‐state results recently reported in...

We report on the noncovalent interactions of nanodiamond carboxylic acids derived from adamantane, diamantane, and triamantane with β- γ-cyclodextrins. The water solubility nanodiamonds was increased by attaching an aromatic dicarboxylic acid via peptide coupling. Isothermal titration calorimetry experiments were performed to determine thermodynamic parameters (Ka, ΔH, ΔG ΔS) for host–guest inclusion. stoichiometry complexes is invariably 1 : 1. It found that Ka, ΔH inclusion increase larger...

Abstract A broad spectrum of physiological processes is mediated by highly specific noncovalent interactions carbohydrates and proteins. In a recent communication we identified several cyclic hexapeptides in dynamic combinatorial library that interact selectively with high binding constants water. Herein, report detailed investigation the interaction two (Cys‐His‐Cys (which call HisHis) Cys‐Tyr‐Cys TyrTyr)) selection monosaccharides disaccharides aqueous solution. The parallel antiparallel...

In this article we report the preparation and characterization of a peptide-based hydrogel, which possesses characteristic rheological properties, is pH responsive can be functionalized at its thiol function. The tripeptide N-(fluorenyl-9-methoxycarbonyl)-L-Cys(acetamidomethyl)-L-His-L-Cys-OH 1 forms stable supramolecular aggregates in water leading to hydrogels above 1.5 wt%. Rheological analysis hydrogel revealed visco-elastic shear thinning properties samples containing wt% peptide 1....

QM/MM models of the peroxo forms vanadium-containing haloperoxidases (VHPOs) are critically assessed in terms active site geometries, hydrogen bonds within site, isotropic and anisotropic 51V NMR chemical shifts, TD-DFT excitation energies. The geometric stability protein is comparable to respective native forms, as indicated by low standard deviations bond lengths across a number local minima sampled along MD trajectories. There significant calculated upfield shift δ(51V) upon formation...

Abstract An artificial nucleoside surrogate with 1 H ‐imidazo[4,5‐ f ][1,10]phenanthroline ( P ) acting as an aglycone has been introduced into DNA oligonucleotide duplexes. This can act a bidentate ligand, and so is useful in the context of metal‐mediated base pairs. Several duplexes involving hetero pair imidazole have investigated. The stability incorporating respective Ag I ‐mediated pairs strongly depends on sequence context. Quantum mechanical/molecular mechanical (QM/MM) calculations...

Quantum-based refinement utilizes chemical restraints derived from quantum-chemical methods instead of the standard parameterized library-based used in packages. The motivation is twofold: firstly, have potential to be more accurate, and secondly, can easily applied new molecules such as drugs or novel cofactors. Here, a project called Q|R aimed at developing quantum-based biomacromolecules under active development by researchers Shanghai University together with PHENIX developers. central...

Owing to the limited quality of experimental data, such as resolution, atomic model refinement using cryo-EM or crystallographic data is often a challenging task. To make practical, priori information about geometry always used chemical restraints. These restraints originate from standard libraries that are across most structure solution software in field. Geometric derived these suffer at least three limitations. First, they lack novel molecules, ligands. Second, agnostic non-covalent...

The experimental electron density distributions (EDD) in tetrafluorophthalonitrile and tetrafluoroisophthalonitrile have been determined from a multipole refinement with use of accurate X-ray diffraction data collected at 100 K. As hydrogen atoms pose serious challenge to successful based EDD determination due their diminishing power, the absence any such studied complexes is an important advantage. compared theoretical densities resulting high-level ab initio DFT calculations using...

Accurately refining biomacromolecules using a quantum-chemical method is challenging because the cost of calculation scales approximately as nm, where n number atoms and m (≥3) based on quantum choice. This fundamental problem means that calculations become intractable when size system requires more computational resources than are available. In development software package called Q|R, this issue referred to Q|R#1. A divide-and-conquer approach has been developed fragments atomic model into...