Abstract We introduce QM7-X, a comprehensive dataset of 42 physicochemical properties for ≈4.2 million equilibrium and non-equilibrium structures small organic molecules with up to seven non-hydrogen (C, N, O, S, Cl) atoms. To span this fundamentally important region chemical compound space (CCS), QM7-X includes an exhaustive sampling (meta-)stable structures—comprised constitutional/structural isomers stereoisomers, e.g., enantiomers diastereomers (including cis- / trans - conformational...

The GEMS method enables molecular dynamics simulations of large heterogeneous systems at ab initio quality.

Two-dimensional semiconductor materials with puckered structure offer a novel playground to implement nanoscale thermoelectric, electronic, and optoelectronic devices improved functionality. Using combination of approaches compute the electronic phonon band structures Green's function based transport techniques, we address thermoelectric performance phosphorene, arsenene, SnS monolayers. In particular, study influence anisotropy in phononic properties its impact on figure merit ZT. Our...

We combine density-functional tight binding (DFTB) with deep tensor neural networks (DTNN) to maximize the strengths of both approaches in predicting structural, energetic, and vibrational molecular properties. The DTNN is used construct a nonlinear model for localized many-body interatomic repulsive energy, which so far has been treated an atom-pairwise manner DFTB. Substantially improving upon standard DFTB DTNN, resulting DFTB-NNrep yields accurate predictions atomization isomerization...

Predictive modeling of toxicity is a crucial step in the drug discovery pipeline. It can help filter out molecules with high probability failing early stages de novo design. Thus, several machine learning (ML) models have been developed to predict by combining classical ML techniques or deep neural networks well-known molecular representations such as fingerprints 2D graphs. But more natural, accurate representation expected be defined physical 3D space like ab initio methods. Recent studies...

Abstract We here introduce the Aquamarine (AQM) dataset, an extensive quantum-mechanical (QM) dataset that contains structural and electronic information of 59,783 low-and high-energy conformers 1,653 molecules with a total number atoms ranging from 2 to 92 (mean: 50.9), containing up 54 28.2) non-hydrogen atoms. To gain insights into solvent effects as well collective dispersion interactions for drug-like molecules, we have performed QM calculations supplemented treatment many-body (MBD)...

Machine Learning Force Fields (MLFFs) promise to enable general molecular simulations that can simultaneously achieve efficiency, accuracy, transferability, and scalability for diverse molecules, materials, hybrid interfaces. A key step toward this goal has been made with the GEMS approach biomolecular dynamics [Sci. Adv. 10, eadn4397 (2024)]. This work introduces SO3LR method integrates fast stable SO3krates neural network semi-local interactions universal pairwise force fields designed...

Machine Learning Force Fields (MLFFs) promise to enable general molecular simulations that can simultaneously achieve efficiency, accuracy, transferability, and scalability for diverse molecules, materials, hybrid interfaces. A key step toward this goal has been made with the GEMS approach biomolecular dynamics [Sci. Adv. 10, eadn4397 (2024)]. This work introduces SO3LR method integrates fast stable SO3krates neural network semi-local interactions universal pairwise force fields designed...

Computer-driven molecular design combines the principles of chemistry, physics, and artificial intelligence to identify chemical compounds with tailored properties. While quantum-mechanical (QM) methods, coupled machine learning, already offer a direct mapping from 3D structures their properties, effective methodologies for inverse in space remain elusive. We address this challenge by demonstrating possibility parametrizing finite set QM Our proof-of-concept implementation achieves an...

By employing a mechanically controllable break junction technique, we have realized an ideal single molecular linear actuator based on dithienylethene (DTE) architecture, which undergoes reversible photothermal isomerization when subjected to UV irradiation under ambient conditions. As result, open form (compressed, OFF) and closed (elongated, ON) of dithienylethene-based junctions are achieved. Interestingly, the mechanical actuation is achieved without changing conductance around Fermi...

Predicting the binding affinity of ligand molecules to protein pockets is a key step in drug design pipeline. The flexibility ligand-pocket motifs arises from wide range attractive and repulsive electronic interactions invoked upon binding. Accurately accounting for all these on equal footing requires robust quantum-mechanical (QM) benchmarks, which are scarce systems. In addition, puzzling disagreement between ``gold standard'' Coupled Cluster (CCSD(T)) Quantum Monte Carlo (QMC) methods...

Biomolecular simulations have been paramount in advancing our understanding of the complex dynamics biological systems. They played a crucial role various applications, including drug discovery and molecular characterization virus-host interactions. Despite their success, biomolecular face inherent challenges due to multiscale nature processes, which involve intricate interactions across wide range length time scales. All-atom (AA-MD) provides detailed insights at atomistic resolution, yet...

Abstract We introduce the MORE-Q dataset, a quantum-mechanical (QM) dataset encompassing structural and electronic data of non-covalent molecular sensors formed by combining 18 mucin-derived olfactorial receptors with 102 body odor volatilome (BOV) molecules. To have better understanding their intra- inter-molecular interactions, we performed accurate QM calculations in different stages sensor design and, accordingly, splits into three subsets: i) MORE-Q-G1: BOV molecules, ii) MORE-Q-G2:...

Abstract The mechanical response of patterned graphene nanoribbons (GNRs) with a width less than 100 nm was studied in - situ using quantitative tensile testing transmission electron microscope (TEM). A high degree crystallinity confirmed for before and after the experiment by selected area diffraction (SAED) patterns. However, maximum local true strain determined to be only about 3%. simultaneously recorded low-loss energy loss spectrum (EELS) on stretched did not reveal any bandgap...

The rational design of molecules with targeted quantum-mechanical (QM) properties requires an advanced understanding the structure-property/property-property relationships (SPR/PPR) that exist across chemical compound space (CCS). In this work, we analyze these fundamental in sector CCS spanned by small (primarily organic) using recently developed QM7-X dataset, a systematic, extensive, and tightly converged collection 42 QM corresponding to ≈4.2M equilibrium non-equilibrium molecular...

Mass transport through graphene is receiving increasing attention due to the potential for molecular sieving. Experimental studies are mostly limited translocation of protons, ions, and water molecules, results larger molecules rare. Here, we perform controlled radical polymerization with surface-anchored self-assembled initiator monolayer in a monomer solution single-layer separating from monomer. We demonstrate that neutral monomers able pass (via native defects) increase defects ratio...

We develop a quantum embedding method that enables accurate and efficient treatment of interactions between molecules an environment, while explicitly including many-body correlations. The molecule is composed classical nuclei electrons, whereas the environment modeled via charged harmonic oscillators. construct general Hamiltonian introduce variational Ansatz for correlated ground state fully interacting molecule-environment system. This wave function optimized Monte Carlo energy...

Copper ions play a major role in biological processes. Abnormal Cu2+ concentrations are associated with various diseases, hence, can be used as diagnostic target. Monitoring copper ion is currently performed by non-portable, expensive and complicated to use equipment. We present label free highly sensitive electrochemical ion-detecting biosensor based on Gly-Gly-His tripeptide layer that chelate ions. The proposed sensing mechanism the chelation results conformational changes peptide forms...

Novel two-dimensional (2D) materials show unusual physical properties which combined with strain engineering open up the possibility of new potential device applications in nanoelectronics. In particular, transport have been found to be very sensitive applied strain. present work, using a density-functional based tight-binding (DFTB) method combination Green's function (GF) approaches, we address effect setup (contact-device(scattering)-contact regions) on electron and phonon materials,...

With the advances in fabrication of materials with feature sizes at order nanometers, it has been possible to alter their thermal transport properties dramatically.Miniaturization device size increases power density general, hence faster electronics require better transport, whereas thermoelectric applications opposite.Such diverse needs bring new challenges for material design.Shrinkage length scales also changed experimental and theoretical methods study transport.Unsurprisingly, novel...

Molecular dynamics (MD) simulations allow atomistic insights into chemical and biological processes. Accurate MD require computationally demanding quantum-mechanical calculations, being practically limited to short timescales few atoms. For larger systems, efficient, but much less reliable empirical force fields are used. Recently, machine learned (MLFFs) emerged as an alternative means execute simulations, offering similar accuracy ab initio methods at orders-of-magnitude speedup. Until...

Abstract The integrity of phonon transport properties large graphene (linear and curved) grain boundaries (GBs) is investigated under the influence structural dynamical disorder. To do this, density functional tight‐binding (DFTB) method combined with atomistic Green's function technique. results show that curved GBs have lower thermal conductance than linear GBs. Its magnitude depends on length curvature out‐of‐plane distortions at boundary, having stronger latter one. Moreover, it found by...

Asymmetric MoS₂nanoribbons display thermal rectification the magnitude of which sensitively depends on their transversal size and localization degree vibrational modes.