Planning chemical syntheses with deep neural networks and symbolic AI.

To plan the syntheses of small organic molecules, chemists use retrosynthesis, a problem-solving technique in which target molecules are recursively transformed into increasingly simpler precursors. Computer-aided retrosynthesis would be a valuable tool but at present it is slow and provides results of unsatisfactory quality. Here we use Monte Carlo tree search and symbolic artificial intelligence (AI) to discover retrosynthetic routes. We combined Monte Carlo tree search with an expansion policy network that guides the search, and a filter network to pre-select the most promising retrosynthetic steps. These deep neural networks were trained on essentially all reactions ever published in organic chemistry. Our system solves for almost twice as many molecules, thirty times faster than the traditional computer-aided search method, which is based on extracted rules and hand-designed heuristics. In a double-blind AB test, chemists on average considered our computer-generated routes to be equivalent to reported literature routes.

Incidence and burden of 671 injuries in professional women footballers: time to focus on context-specific injury risk reduction strategies.

This study investigated the extent of injury incidence and burden in a professional women football team of the Scottish Women's Premier League during two seasons. All injuries causing time-loss or required medical attention were recorded prospectively. A total of 671 injuries, 570 requiring medical attention and 101 causing time-loss were recorded in 41 players. Injuries occurring with National Team resulted in 12% of the club's international players' lay-off. Overall injury incidence was 11.1/1000-hours and burden was 368.9 days/1000-hours. Injury incidence (23.9/1000-hours vs 8.2/1000-hours) and burden (1049.8 days/1000-hours vs 215.1 days/1000-hours) were higher for match compared to training. Foremost mechanism of match injury burden was indirect-contact, which was different than the non-contact predominantly observed for training injury burden. Injury incidence, burden and patterns differed between training, match and playing positions. Tailoring injury-risk reduction strategies considering context, circumstances and playing position deserve consideration to enhance player's injury resilience in professional women footballers.

Yoink: An interaction-based partitioning API.

Herein, we describe the implementation details of our interaction-based partitioning API (application programming interface) called Yoink for QM/MM modeling and fragment-based quantum chemistry studies. Interactions are detected by computing density descriptors such as reduced density gradient, density overlap regions indicator, and single exponential decay detector. Only molecules having an interaction with a user-definable QM core are added to the QM region of a hybrid QM/MM calculation. Moreover, a set of molecule pairs having density-based interactions within a molecular system can be computed in Yoink, and an interaction graph can then be constructed. Standard graph clustering methods can then be applied to construct fragments for further quantum chemical calculations. The Yoink API is licensed under Apache 2.0 and can be accessed via yoink.wallerlab.org. © 2018 Wiley Periodicals, Inc.

Rational Density Functional Selection Using Game Theory.

Theoretical chemistry has a paradox of choice due to the availability of a myriad of density functionals and basis sets. Traditionally, a particular density functional is chosen on the basis of the level of user expertise (i.e., subjective experiences). Herein we circumvent the user-centric selection procedure by describing a novel approach for objectively selecting a particular functional for a given application. We achieve this by employing game theory to identify optimal functional/basis set combinations. A three-player (accuracy, complexity, and similarity) game is devised, through which Nash equilibrium solutions can be obtained. This approach has the advantage that results can be systematically improved by enlarging the underlying knowledge base, and the deterministic selection procedure mathematically justifies the density functional and basis set selections.

Modelling Chemical Reasoning to Predict and Invent Reactions.

The ability to reason beyond established knowledge allows organic chemists to solve synthetic problems and invent novel transformations. Herein, we propose a model that mimics chemical reasoning, and formalises reaction prediction as finding missing links in a knowledge graph. We have constructed a knowledge graph containing 14.4 million molecules and 8.2 million binary reactions, which represents the bulk of all chemical reactions ever published in the scientific literature. Our model outperforms a rule-based expert system in the reaction prediction task for 180 000 randomly selected binary reactions. The data-driven model generalises even beyond known reaction types, and is thus capable of effectively (re-)discovering novel transformations (even including transition metal-catalysed reactions). Our model enables computers to infer hypotheses about reactivity and reactions by only considering the intrinsic local structure of the graph and because each single reaction prediction is typically achieved in a sub-second time frame, the model can be used as a high-throughput generator of reaction hypotheses for reaction discovery.

Neural-Symbolic Machine Learning for Retrosynthesis and Reaction Prediction.

Reaction prediction and retrosynthesis are the cornerstones of organic chemistry. Rule-based expert systems have been the most widespread approach to computationally solve these two related challenges to date. However, reaction rules often fail because they ignore the molecular context, which leads to reactivity conflicts. Herein, we report that deep neural networks can learn to resolve reactivity conflicts and to prioritize the most suitable transformation rules. We show that by training our model on 3.5 million reactions taken from the collective published knowledge of the entire discipline of chemistry, our model exhibits a top10-accuracy of 95 % in retrosynthesis and 97 % for reaction prediction on a validation set of almost 1 million reactions.

Sequence-Dependent Duplex Stabilization upon Formation of a Metal-Mediated Base Pair.

An artificial nucleoside surrogate with 1H-imidazo[4,5-f][1,10]phenanthroline (P) acting as an aglycone has been introduced into DNA oligonucleotide duplexes. This nucleoside surrogate can act as a bidentate ligand, and so is useful in the context of metal-mediated base pairs. Several duplexes involving a hetero base pair with an imidazole nucleoside have been investigated. The stability of DNA duplexes incorporating the respective Ag(I) -mediated base pairs strongly depends on the sequence context. Quantum mechanical/molecular mechanical (QM/MM) calculations have been performed in order to gain insight into the factors determining this sequence dependence. The results indicated that, in addition to the stabilizing effect that results from the formation of coordinative bonds, destabilizing effects may occur when the artificial base pair does not fit optimally into the surrounding B-DNA duplex.

pH response and molecular recognition in a low molecular weight peptide hydrogel.

In this article we report the preparation and characterization of a peptide-based hydrogel, which possesses characteristic rheological properties, is pH responsive and 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 of the hydrogel revealed visco-elastic and shear thinning properties of samples containing 1.5 wt% of peptide 1. The hydrogel reversibly responds to pH changes. Below and above pH 6, electrostatic repulsion of the peptide results in a weakening of the three-dimensional gel network. Based on atomic force microscopy, small angle X-ray scattering and molecular dynamics simulations, it is proposed that the peptide assembles into nanostructures that tend to entangle at higher concentrations in water. The development of functional materials based on the peptide assemblies was possible via thiol-ene-click chemistry of the free thiol function at the C-terminal cysteine unit. As a proof of concept, the functionalization with adamantyl units to give 1-Ad was shown by molecular recognition of ß-cyclodextrin vesicles. These vesicles were used as supramolecular cross-linkers of the assemblies of peptide 1 mixed with peptide 1-Ad leading to gel networks at a reduced peptide concentration.

New matrix metalloproteinase inhibitors based on γ-fluorinated α-aminocarboxylic and α-aminohydroxamic acids.

Matrix metalloproteinases (MMPs) are involved in a number of physiological as well as pathological processes such as atherosclerosis and tumorigenesis, where an up-regulation of MMPs is predominant. Fluorinated analogues of the hydroxamate-based non-peptidic broad-spectrum MMP inhibitor (MMPI) CGS 27023A were synthesized and inhibition potencies for MMP-2 and MMP-9 in the nanomolar range were measured using fluorimetric in vitro assays. The inhibition potencies of the herein reported fluorinated MMPIs were comparable or even superior in some cases to their non-fluorinated analogues. In contrast to the lead structure, both enantiomers of fluorinated MMPs were almost equally potent. Modelling studies suggest that the core α-amino hydroxamic acid residues appear to influence the relative potencies via specific inhibitor-peptidase interactions, including short fluorine-hydrogen contacts, within the enzyme's pockets. The binding of the essential hydroxamate group to the zinc ion is rather unaffected by the rest of the molecule. In contrast, the corresponding α-aminocarboxylic acid derivatives are 10(3) times less potent or were even inactive.

Peptide-based carbohydrate receptors.

A broad spectrum of physiological processes is mediated by highly specific noncovalent interactions of carbohydrates and proteins. In a recent communication we identified several cyclic hexapeptides in a dynamic combinatorial library that interact selectively with carbohydrates with high binding constants in water. Herein, we report a detailed investigation of the noncovalent interaction of two cyclic hexapeptides (Cys-His-Cys (which we call HisHis) and Cys-Tyr-Cys (which we call TyrTyr)) with a selection of monosaccharides and disaccharides in aqueous solution. The parallel and antiparallel isomers of HisHis or TyrTyr were synthesized separately, and their interaction with monosaccharides and disaccharides in aqueous solution was studied by isothermal titration calorimetry, NMR spectroscopic titrations, and circular dichroism spectroscopy. From these measurements, we identified particularly stable complexes (Ka> 1000 M(-1)) of the parallel isomer of HisHis with N-acetylneuraminic acid and with methyl-a-d-galactopyranoside as well as of both isomers of TyrTyr with trehalose. To gain further insight into the structure of the peptide­carbohydrate complexes, structure prediction was performed using quantum chemical methods. The calculations confirm the selectivity observed in the experiments and indicate the formation of multiple intermolecular hydrogen bonds in the most stable complexes.

A density-based adaptive quantum mechanical/molecular mechanical method.

We present a density-based adaptive quantum mechanical/molecular mechanical (DBA-QM/MM) method, whereby molecules can switch layers from the QM to the MM region and vice versa. The adaptive partitioning of the molecular system ensures that the layer assignment can change during the optimization procedure, that is, on the fly. The switch from a QM molecule to a MM molecule is determined if there is an absence of noncovalent interactions to any atom of the QM core region. The presence/absence of noncovalent interactions is determined by analysis of the reduced density gradient. Therefore, the location of the QM/MM boundary is based on physical arguments, and this neatly removes some empiricism inherent in previous adaptive QM/MM partitioning schemes. The DBA-QM/MM method is validated by using a water-in-water setup and an explicitly solvated L-alanyl-L-alanine dipeptide.

AQuaRef: Machine learning accelerated quantum refinement of protein structures.

Cryo-EM and X-ray crystallography provide crucial experimental data for obtaining atomic-detail models of biomacromolecules. Refining these models relies on library-based stereochemical restraints, which, in addition to being limited to known chemical entities, do not include meaningful noncovalent interactions relying solely on nonbonded repulsions. Quantum mechanical (QM) calculations could alleviate these issues but are too expensive for large molecules. We present a novel AI-enabled Quantum Refinement (AQuaRef) based on AIMNet2 neural network potential mimicking QM at substantially lower computational costs. By refining 41 cryo-EM and 30 X-ray structures, we show that this approach yields atomic models with superior geometric quality compared to standard techniques, while maintaining an equal or better fit to experimental data.

Revealing noncovalent interactions in quantum crystallography: taurine revisited.

The charge density distribution in taurine (2-aminoethane-sulfonic acid) is further studied with the molecular orbital occupation number refinement scheme. The recently proposed NCIPLOT scheme (Johnson et al., J. Am. Chem. Soc. 2010, 132, 6498) is applied to visualize the noncovalent interactions from experimentally refined charge densities. Herein, we demonstrate the evolution of the reduced density gradient isosurface during the charge density refinement process.

JACOB: an enterprise framework for computational chemistry.

Here, we present just a collection of beans (JACOB): an integrated batch-based framework designed for the rapid development of computational chemistry applications. The framework expedites developer productivity by handling the generic infrastructure tier, and can be easily extended by user-specific scientific code. Paradigms from enterprise software engineering were rigorously applied to create a scalable, testable, secure, and robust framework. A centralized web application is used to configure and control the operation of the framework. The application-programming interface provides a set of generic tools for processing large-scale noninteractive jobs (e.g., systematic studies), or for coordinating systems integration (e.g., complex workflows). The code for the JACOB framework is open sourced and is available at: www.wallerlab.org/jacob.

A systematic approach to identify cooperatively bound homotrimers.

A systematic multistage computational procedure is presented to investigate cooperativity within trimeric molecular complexes. It is an alternative to exhaustive or stochastic-based approaches. Trimeric clusters were extracted from known crystal structures and optimized in the gas phase, and subsequently filtered using energetic and RMSD structural cutoffs. Three-body interaction energies were computed for the subset of distinct low-energy trimer conformations. The procedure was validated using a set of 20 molecular crystals taken from the Cambridge Structural Database, and 25% of these strucutres gave rise to gas-phase homotrimers that showed a cooperative binding energy at the BP86-D3(BJ)/def2-SVP//TPSS-D3(BJ)/def2-TZVP level of theory.

Investigating inclusion complexes using quantum chemical methods.

Quantum chemistry has firmly established itself as a reliable method for investigating present-day problems in biological and materials chemistry. Understanding inclusion complexes represents one of the cutting edges of simulation sciences. In this tutorial review, we focus on the role and composition of non-covalent interactions, which are essential when studying inclusion complexes. A selected set of recently developed pragmatic methods used to study inclusion complexes are then surveyed including e.g. dispersion corrected DFT, double-hybrid functionals and spin-component scaled MP2. Finally, three case studies are outlined: (a) endohedral fullerene complexes, (b) buckyball catcher and (c) resorcinarene capsule. These case studies were carefully chosen to help illustrate how one may accurately investigate inclusion complexes, at a modest computational cost, using state-of-the-art quantum chemical methods (67 references).

pH-Switchable ampholytic supramolecular copolymers.

ß-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 in response to pH triggers. The cooperative supramolecular self-assembly process is switched on at a physiologically relevant pH value and can be switched off by increasing or decreasing the pH value.

Nanodiamonds in sugar rings: an experimental and theoretical investigation of cyclodextrin-nanodiamond inclusion complexes.

We report on the noncovalent interactions of nanodiamond carboxylic acids derived from adamantane, diamantane, and triamantane with ß- and γ-cyclodextrins. The water solubility of the nanodiamonds was increased by attaching an aromatic dicarboxylic acid via peptide coupling. Isothermal titration calorimetry experiments were performed to determine the thermodynamic parameters (K(a), ΔH, ΔG and ΔS) for the host-guest inclusion. The stoichiometry of the complexes is invariably 1:1. It was found that K(a), ΔG and ΔH of inclusion increase for larger nanodiamonds. ΔS is generally positive, in particular for the largest nanodiamonds. ß-Cyclodextrin binds all nanodiamonds, γ-cyclodextrin clearly prefers the most bulky nanodiamonds. The interaction of 9-triamantane carboxylic acid shows one of the strongest complexation constants towards γ-cyclodextrin ever reported, K(a) = 5.0 × 10(5) M(-1). In order to gain some insight into the possible structural basis of these inclusion complexes we performed density functional calculations at the B97-D3/def2-TZVPP level of theory.

JACOB: a dynamic database for computational chemistry benchmarking.

JACOB (just a collection of benchmarks) is a database that contains four diverse benchmark studies, which in-turn included 72 data sets, with a total of 122,356 individual results. The database is constructed upon a dynamic web framework that allows users to retrieve data from the database via predefined categories. Additional flexibility is made available via user-defined text-based queries. Requested sets of results are then automatically presented as bar graphs, with parameters of the graphs being controllable via the URL. JACOB is currently available at www.wallerlab.org/jacob.

Assessment of weak intermolecular interactions across QM/MM noncovalent boundaries.

An assessment of a number of quantum mechanical/molecular mechanical (QM/MM) combinations was performed for weak intermolecular interactions across noncovalent QM/MM 'boundaries'. The popular S22 data set, comprising of a number of weak hydrogen-bonded, dispersion-bound and complexes with mixed interactions was used for the assessment. A range of QM methods was combined with a number of popular MM force fields. The single-point interaction energies, at reference geometries, are presented as deviations to accurate CCSD(T)/CBS reference values. This investigation employed both additive and subtractive QM/MM schemes. The density functional has only a negligible effect; the choice of basis set was also negligible in terms of accuracy. The importance of selecting the most appropriate MM force field for accurately describing interactions at the noncovalent 'boundary' region has a dramatic effect on the accuracy.