Featured
-
-
Article
| Open AccessA human-machine interface for automatic exploration of chemical reaction networks
Automated reaction exploration is the key to systematic elucidation of chemical mechanisms. Here, the authors introduce a generally applicable algorithm to steer an automated exploration towards region of interest in chemical reaction space.
- Miguel Steiner
- & Markus Reiher
-
Article
| Open AccessThe role of charge in microdroplet redox chemistry
Redox reactions exhibit different thermodynamics and kinetics in water microdroplets compared to the bulk. Here, the authors use reactive molecular dynamics to show the importance of charged reactive species in explaining reactivity under confinement.
- Joseph P. Heindel
- , R. Allen LaCour
- & Teresa Head-Gordon
-
Article
| Open AccessUnderstanding X-ray absorption in liquid water using triple excitations in multilevel coupled cluster theory
Accurate modeling of X-ray absorption (XA) is necessary to interpret experimental spectra. Here, the authors use coupled cluster models and path-integral ab initio molecular dynamics to provide insights into the XA spectrum of liquid water.
- Sarai Dery Folkestad
- , Alexander C. Paul
- & Henrik Koch
-
Article
| Open AccessComparative quantum-classical dynamics of natural and synthetic molecular rotors show how vibrational synchronization modulates the photoisomerization quantum efficiency
The control of reaction quantum efficiencies is a fundamental photochemical problem. Here the authors use comparative quantum-classical dynamics to reveal that the synchronization of specific vibrations with the reaction coordinate is a key promoting factor.
- Alejandro Blanco-Gonzalez
- , Madushanka Manathunga
- & Massimo Olivucci
-
Article
| Open AccessPolarizability matters in enantio-selection
Polarizability, a property that is closely related to softness in the classic theory of Hard and Soft Acids and Bases (HSAB), has been largely overlooked in connecting with enantio-selection in the past. Here, the authors show local polarizability-based electronic effects can rationalize a wide range of stereochemical outcomes in widely-known asymmetric catalytic reactions.
- Fumin Chen
- , Yu Chen
- & Xiangyou Xing
-
Article
| Open AccessProspective de novo drug design with deep interactome learning
The use of data-driven generative models for drug design is challenging due to the scarcity of data. Here, the authors introduce a “zero-shot" generative deep model to enable the generation of molecules by both structure- and ligand-based drug design and apply it to design PPARγ agonists with desired properties.
- Kenneth Atz
- , Leandro Cotos
- & Gisbert Schneider
-
Article
| Open Access3D molecular generative framework for interaction-guided drug design
Designing a molecule that favorably binds to a protein pocket is a keystone of drug discovery. Zhung et al. devise DeepICL, which leverages the generalizable features of non-covalent protein-ligand interactions on a 3D molecular generative model, improving the quality of AI-designed molecules.
- Wonho Zhung
- , Hyeongwoo Kim
- & Woo Youn Kim
-
Article
| Open AccessDeveloping a machine learning model for accurate nucleoside hydrogels prediction based on descriptors
Supramolecular nucleoside-based hydrogels have potential in biomedical applications, but there is no model to predict what nucleoside derivatives will form hydrogels. Here, the authors report a machine learning model to predict the ability of nucleoside derivatives to form hydrogels.
- Weiqi Li
- , Yinghui Wen
- & Hang Zhao
-
Article
| Open AccessDynamics of the charge transfer to solvent process in aqueous iodide
Solvated electrons can be formed through photo-induced charge-transfer-to-solvent electronic states of halide ions in water. Here, the authors use machine learning accelerated molecular dynamics simulations to follow the evolution of these states for aqueous iodide in detail.
- Jinggang Lan
- , Majed Chergui
- & Alfredo Pasquarello
-
Article
| Open AccessActivation and friction in enzymatic loop opening and closing dynamics
Enzymes present loops around active sites whose closing and opening dynamics are essential for its activity. Here the authors unveil the mechanism governing loop motion, showing that it involves an activated conformational rearrangement around a couple of torsional angles taking place under the strong friction exerted by the rest of loop torsions.
- Kirill Zinovjev
- , Paul Guénon
- & Iñaki Tuñón
-
Article
| Open AccessAqueous pyruvate partly dissociates under deep ultraviolet irradiation but is resilient to near ultraviolet excitation
The photochemistry of the pyruvate anion plays an important role in the Earth’s atmosphere and aqueous environments. Here, the authors show that excitation of aqueous pyruvate by 200 nm light leads to decarboxylation with a quantum efficiency of 20%, while excitation by 340 nm light does not cause a reaction.
- Jan Thøgersen
- , Fani Madzharova
- & Frank Jensen
-
Article
| Open AccessDesign of target specific peptide inhibitors using generative deep learning and molecular dynamics simulations
Here the authors report a computational approach which integrates deep learning and structural modelling to design target-specific peptides. They apply this to β-catenin and NF-κB essential modulator, resulting in improved binding, highlighting the efficacy of this strategy.
- Sijie Chen
- , Tong Lin
- & Xiaolin Cheng
-
Article
| Open AccessIterative design of training data to control intricate enzymatic reaction networks
Kinetic modeling of in vitro enzymatic reaction networks (ERNs) is severely hampered by the lack of training data. Here, authors introduce a methodology that combines an active learning-like approach and flow chemistry to create optimized datasets for an intricate ERN.
- Bob van Sluijs
- , Tao Zhou
- & Wilhelm T. S. Huck
-
Article
| Open AccessOverlay databank unlocks data-driven analyses of biomolecules for all
In this work, the authors report NMR lipids Databank to promote decentralised sharing of biomolecular molecular dynamics (MD) simulation data with an overlay design. Programmatic access enables analyses of rare phenomena and advances the training of machine learning models.
- Anne M. Kiirikki
- , Hanne S. Antila
- & O. H. Samuli Ollila
-
Article
| Open AccessSQM2.20: Semiempirical quantum-mechanical scoring function yields DFT-quality protein–ligand binding affinity predictions in minutes
The paper presents the universal QM-based scoring function that accurately and rapidly predicts protein-ligand binding affinities, outperforming current computational tools. This is demonstrated on the PL-REX experimental benchmark dataset.
- Adam Pecina
- , Jindřich Fanfrlík
- & Jan Řezáč
-
Article
| Open AccessThe nature of carotenoid S* state and its role in the nonphotochemical quenching of plants
Plant Light Harvesting complexes adjust to light conditions via a quenching mechanism involving carotenoids. The authors use computational simulations to reveal how carotenoids’ quenching capacity is tuned by conformational changes of the complex.
- Davide Accomasso
- , Giacomo Londi
- & Benedetta Mennucci
-
Article
| Open AccessDiffusion-based generative AI for exploring transition states from 2D molecular graphs
The exploration of transition state (TS) geometries is crucial for elucidating chemical reaction mechanisms and modelling their kinetics. Here, authors propose a generative AI approach to predict TS geometries just from 2D molecular graphs of a reaction.
- Seonghwan Kim
- , Jeheon Woo
- & Woo Youn Kim
-
Article
| Open AccessMonovalent metal ion binding promotes the first transesterification reaction in the spliceosome
Hybrid QM/MM molecular dynamics simulations reveal that the kinetics and thermodynamics of the first splicing step are regulated by a K+ ion that facilitates optimal positioning of reactive moieties.
- Jana Aupič
- , Jure Borišek
- & Alessandra Magistrato
-
Article
| Open AccessElucidation of the structural basis for ligand binding and translocation in conserved insect odorant receptor co-receptors
Insects rely on olfaction for behavior control. Recent structural studies of receptors provide insight into ligand binding. Here, the authors identify dynamic binding mechanism to Orco, explaining its high selectivity with insights in compound screening.
- Jody Pacalon
- , Guillaume Audic
- & Jérémie Topin
-
Article
| Open AccessDocking for EP4R antagonists active against inflammatory pain
Non-addictive treatments for pain are much needed. Here, the authors identify in vivo active leads for inflammatory pain using large library docking against the EP4 prostaglandin receptor.
- Stefan Gahbauer
- , Chelsea DeLeon
- & Brian K. Shoichet
-
Article
| Open AccessThe carbonyl-lock mechanism underlying non-aromatic fluorescence in biological matter
Recent experimental evidence shows a new type of intrinsic fluorescence in biomolecules void of aromatic chemical compounds whose origin is unclear. Here, the authors use non-adiabatic AIMD simulations to show a potential carbonyl-lock mechanism originating this phenomenon.
- Gonzalo Díaz Mirón
- , Jonathan A. Semelak
- & Uriel N. Morzan
-
Article
| Open AccessThermal dependence of the hydrated proton and optimal proton transfer in the protonated water hexamer
Water’s pivotal role is tied to the quantum nature of its hydrogen bond dynamics. Here, the authors investigate the thermal behavior of the protonated water hexamer through accurate path integral molecular dynamics, revealing that near-room temperature conditions are optimal for proton transfer.
- Félix Mouhat
- , Matteo Peria
- & Michele Casula
-
Article
| Open AccessHighly active, ultra-low loading single-atom iron catalysts for catalytic transfer hydrogenation
Highly effective and selective noble metal-free catalysts continue to attract significant attention but require reaction specific tuning. Here, the authors fabricate a single-atom iron catalyst at low loading, which shows excellent transfer hydrogenation performance even at low reaction temperatures.
- Zhidong An
- , Piaoping Yang
- & Dionisios G. Vlachos
-
Comment
| Open AccessLimitations of representation learning in small molecule property prediction
Machine learning is a powerful tool for the study and design of molecules. Here the authors comment a recent publication in Nature Communications which highlights the challenges of different molecular representations for data-driven property predictions.
- Ana Laura Dias
- , Latimah Bustillo
- & Tiago Rodrigues
-
Article
| Open AccessChallenging an old paradigm by demonstrating transition metal-like chemistry at a neutral nonmetal center
The scope of and knowledge about the chemistry of nonmetal-adducts remains very limited. Here, the authors describe nonmetal adducts of the phosphorus center of terminal phosphinidene complexes using classical C- and N-ligands from metal coordination chemistry.
- David Biskup
- , Gregor Schnakenburg
- & Rainer K. Streubel
-
Article
| Open AccessStructural basis of dimerization of chemokine receptors CCR5 and CXCR4
Here, authors report chemokine receptors structures obtained using coarse-grained metadynamics. CCR5 and CXCR4 homo- and heterodimers differ in the conformations of ligand binding sites and of the G protein interaction interface, suggesting structural basis for the rational design of biased ligands.
- Daniele Di Marino
- , Paolo Conflitti
- & Vittorio Limongelli
-
Article
| Open AccessControlling piezoresistance in single molecules through the isomerisation of bullvalenes
The quest for miniaturisation of electromechanical nanosystems requires the use of single molecules as active components. Here, the authors develop a piezoresistor based on a single bullvalene molecule that changes its shape by a Cope rearrangement.
- Jeffrey R. Reimers
- , Tiexin Li
- & Nadim Darwish
-
Article
| Open AccessMinimizing non-radiative decay in molecular aggregates through control of excitonic coupling
Exciton delocalization in molecular aggregates is suggested to counteract the Energy Gap Law. Here, authors reveal the underlying physical picture and find the optimal excitonic coupling that minimizes nonradiative decay by nearly exact simulations.
- Yuanheng Wang
- , Jiajun Ren
- & Zhigang Shuai
-
Article
| Open AccessMacrocyclization of linear molecules by deep learning to facilitate macrocyclic drug candidates discovery
Macrocyclization of bioactive acyclic molecules provides a potential avenue to yield novel chemical scaffolds with improved pharmacological properties. Here, the authors propose a deep learning based macrocyclization method to generate diverse macrocycles from a given acyclic molecule.
- Yanyan Diao
- , Dandan Liu
- & Honglin Li
-
Article
| Open AccessActinide inverse trans influence versus cooperative pushing from below and multi-center bonding
Actinide-ligand bonds with high multiplicities remain poorly understood. Here, the authors investigate covalency in actinide complexes and identify terminal O and N ligands that are triply to quadruply bonded to the actinide, facilitated by electrostatic, steric, and covalent interactions.
- Laura C. Motta
- & Jochen Autschbach
-
Article
| Open AccessPotential-dependent transition of reaction mechanisms for oxygen evolution on layered double hydroxides
The mechanisms for oxygen evolution reaction on layered double hydroxides remain controversial. Here, the authors use a computational methodology by combining grand-canonical methods and microkinetic modeling to unravel the potential-dependent transitions mechanisms for electrochemical reactions.
- Zeyu Wang
- , William A. Goddard III
- & Hai Xiao
-
Matters Arising
| Open AccessReply to: On the existence of collective interactions reinforcing the metal-ligand bond in organometallic compounds
- Vojtech Šadek
- , Shahin Sowlati-Hashjin
- & Cina Foroutan-Nejad
-
Matters Arising
| Open AccessOn the existence of collective interactions reinforcing the metal-ligand bond in organometallic compounds
- Jordi Poater
- , Pascal Vermeeren
- & Miquel Solà
-
Article
| Open AccessDesigning main-group catalysts for low-temperature methane combustion by ozone
Automated reaction route mapping is used to design catalysts for low-temperature CH4 combustion with ozone. A suitable proton-type zeolite catalyst with Brønsted acid sites was predicted and shown to have superior performance in CH4 combustion.
- Shunsaku Yasumura
- , Kenichiro Saita
- & Ken-ichi Shimizu
-
Article
| Open AccessEfficient interatomic descriptors for accurate machine learning force fields of extended molecules
Accurate description of non-local interactions represents a challenge for machine learning force fields. Here, authors develop linearly scaling global descriptors and analyse the non-local interatomic features that contribute to accurate predictions.
- Adil Kabylda
- , Valentin Vassilev-Galindo
- & Alexandre Tkatchenko
-
Article
| Open AccessRealistic phase diagram of water from “first principles” data-driven quantum simulations
The molecular modelling of water has been a long sought-after goal in computational sciences for more than 50 years. Here, the authors show that the data-driven many-body MB-pol potential can provide a realistic representation of the phase diagram of water.
- Sigbjørn Løland Bore
- & Francesco Paesani
-
Article
| Open AccessMachine learning-enabled exploration of the electrochemical stability of real-scale metallic nanoparticles
Surface Pourbaix diagrams are critical to understanding the stability of nanomaterials. Here, the authors develop a bond-type embedded crystal graph convolutional neural network model and construct reliable Pourbaix diagrams for real-scale nanoparticles.
- Kihoon Bang
- , Doosun Hong
- & Hyuck Mo Lee
-
Article
| Open AccessDeducing subnanometer cluster size and shape distributions of heterogeneous supported catalysts
IR spectra are great for characterizing single-crystals and large nanoparticles, but not for highly dispersed heterogeneous catalysts made up of single-atoms and ultra-small clusters. To solve this, the authors developed a method to generate synthetic IR spectra using data-based approaches and physics-driven surrogate models.
- Vinson Liao
- , Maximilian Cohen
- & Dionisios G. Vlachos
-
Article
| Open AccessTowards the ground state of molecules via diffusion Monte Carlo on neural networks
An accurate ab initio calculation of molecules is fundamental to chemical and physical sciences. Here, the authors integrate a neural-network wavefunction into the fixed-node diffusion Monte Carlo, resulting in accurate calculations of a diverse range of systems, offering insights into complex many-body electronic wave functions.
- Weiluo Ren
- , Weizhong Fu
- & Ji Chen
-
Article
| Open AccessPotent and selective covalent inhibition of the papain-like protease from SARS-CoV-2
The development of direct-acting antivirals to combat COVID-19 remains an important goal. Here the authors design covalent inhibitors that target the papain-like protease from SARS-CoV-2. The most promising inhibitor blocks viral replication in mammalian cells.
- Brian C. Sanders
- , Suman Pokhrel
- & Jerry M. Parks
-
Article
| Open AccessNuclear quantum effects on zeolite proton hopping kinetics explored with machine learning potentials and path integral molecular dynamics
The quantum properties of hydrogen atoms in zeolite-catalyzed reactions are generally neglected due to high computational costs. Here, the authors leverage machine learning to derive accurate quantum kinetics for proton transfer reactions in heterogeneous catalysis.
- Massimo Bocus
- , Ruben Goeminne
- & Veronique Van Speybroeck
-
Article
| Open AccessSynergistic promotions between CO2 capture and in-situ conversion on Ni-CaO composite catalyst
The integrated CO2 capture and conversion (iCCC) technology has been booming for carbon neutrality. Here the authors optimized the Ni–CaO composite catalyst to promote iCCC involving consecutive high-temperature Calcium-looping and dry reforming of methane and illustrated their synergistic promotions at the suitable catalyst interface.
- Bin Shao
- , Zhi-Qiang Wang
- & Jun Hu
-
Article
| Open AccessAn unexpected synthesis of azepinone derivatives through a metal-free photochemical cascade reaction
Photochemical nitrene transfer offers a green avenue for heterocyclic syntheses. Here, the authors developed a metal-free, visible light-mediated cascade reaction for the preparation of azepinone derivatives.
- Lina Song
- , Xianhai Tian
- & A. Stephen K. Hashmi
-
Article
| Open AccessExtending density functional theory with near chemical accuracy beyond pure water
DFT simulations may be inaccurate in modeling aqueous systems, with results depending on the choice of the exchange-correlation functional. Here, the authors present an integrative method called HF-r2SCAN-DC4 that provides near chemical accuracy in electronic structure information not only for pure water but also for molecules dissolved in it
- Suhwan Song
- , Stefan Vuckovic
- & Kieron Burke
-
Article
| Open AccessElectronic excited states in deep variational Monte Carlo
Deep neural networks can learn and represent nearly exact electronic ground states. Here, the authors advance this approach to excited states, achieving high accuracy across a range of atoms and molecules, opening up the possibility to model many excited-state processes.
- M. T. Entwistle
- , Z. Schätzle
- & F. Noé
-
Article
| Open AccessData-driven design of molecular nanomagnets
Three decades of research in molecular nanomagnets have enabled the preparation of compounds displaying magnetic memory at liquid nitrogen temperature. Here, the authors provide an innovative framework for the design of molecular magnets based on data mining, and develop an interactive dashboard to visualize the dataset.
- Yan Duan
- , Lorena E. Rosaleny
- & Alejandro Gaita-Ariño
-
Article
| Open AccessObservation of a transient intermediate in the ultrafast relaxation dynamics of the excess electron in strong-field-ionized liquid water
A unified picture of the electronic relaxation dynamics of ionized liquid water remains elusive despite decades of study. Here, the authors use few-cycle optical pump-probe spectroscopy and ab initio quantum dynamics to unambiguously identify a new transient intermediate in the relaxation pathway.
- Pei Jiang Low
- , Weibin Chu
- & Zhi-Heng Loh
-
Article
| Open AccessDeepPROTACs is a deep learning-based targeted degradation predictor for PROTACs
The rational design of PROTACs is difficult due to their obscure structure-activity relationship. Here the authors present a deep neural network model - DeepPROTACs - for predicting the degradation capacity of a proposed PROTAC molecule.
- Fenglei Li
- , Qiaoyu Hu
- & Fang Bai
-
Article
| Open AccessCharge-separation driven mechanism via acylium ion intermediate migration during catalytic carbonylation in mordenite zeolite
The tremendous application of carbonylation reaction requires the elaborate explanation to reaction mechanism. Here the authors propose a charge-separation driven mechanism of methyl acetate formation via acylium ion intermediate in mordenite zeolite by an integrated reaction/diffusion kinetics model during the dimethyl ether carbonylation.
- Wei Chen
- , Karolina A. Tarach
- & Anmin Zheng