computational-chemistry

Nature Communications, Published online: 20 June 2026; doi:10.1038/s41467-026-74537-4 Solving the full-dimensional vibrational Schrödinger equation remains challenging for polyatomic molecules. The authors present an efficient neural-network approach for accurate vibrational energy calculations for molecules with more than ten atoms.

In computational quantum chemistry, the Born-Oppenheimer approximation (BOA) is used to determine potential energy surfaces (PES) for electronic states of molecules. It is standard practise to identify local minima on a PES with molecular structures. Molecular binding energies are identified with the difference in energy between these minima and the energies of the isolated atoms of which the mol…

An international research team led by a Korean scientist has succeeded in designing large-scale protein structures that faithfully replicate the self-assembly principles found in naturally occurring viruses, using artificial intelligence (AI).

Scientific Reports, Published online: 18 June 2026; doi:10.1038/s41598-026-58227-1 Energy harvesting potential of LiHfIrZ (Z = Si, Ge) quaternary Heusler alloys: a first-principles computational study

Nature Communications, Published online: 17 June 2026; doi:10.1038/s41467-026-73904-5 Metal-organic frameworks are candidates for carbon capture. Here the authors introduce a computational workflow to investigate their CO2 capture performance considering molecular diffusivity and identify promising water-tolerant systems.

Quantitative Structure-Property Relationship (QSPR) modelling provides an efficient computational framework for predicting physicochemical properties of drug molecules when experimental data are limited. In this study, we investigate the predictive capability of degree-based topological indices (TIs) derived from SMILES (Simplified Molecular Input Line Entry System) representations for modelling …

Researchers from the National University of Singapore (NUS) have developed a computation-guided strategy to produce urea more efficiently from carbon dioxide and nitrate. By combining large language models, density functional theory calculations and experiments, the approach identified a cadmium-modified iron oxide catalyst that maintains high urea selectivity at practical current densities. Urea…

Nature Communications, Published online: 13 June 2026; doi:10.1038/s41467-026-74101-0 Predicting transition states of chemical reactions is demanding. A generative flow model in distance geometry space, dubbed TS-DFM, is proposed, achieving higher structural accuracy and better generalizability than Cartesian-based approaches.

Researchers used machine learning to help predict chemical signatures for over 1 billion possible fentanyls, including variants never seen before.

Introduction: Apoptosis resistance in non-small cell lung cancer is frequently sustained by pro-survival Bcl-2 family proteins such as Mcl-1, motivating the search for new Mcl-1 inhibitors from natural products, including diterpenoids from Rabdosia serra . Materials and Methods: An integrated in silico approach was applied to evaluate R. serra diterpenoids as putative Mcl-1 (PDB: 6QFQ) inhibitors…

A new AI model has become so good at predicting how molecules evolve over time that, in the future, it could speed up the costly and time-consuming process of testing new drugs. In the long term, this technology could facilitate the development of medicines and new treatments, as promising drug cand...

A new AI model has become so good at predicting how molecules evolve over time that, in the future, it could speed up the costly and time-consuming process of testing new drugs.

High-resolution phase imaging is vital for nanomanufacturing and bioimaging, but is often limited by the trade-off between resolution and field of view. Scientists in China have developed L2-CPI, a computational phase imaging technique that achieves an arbitrary field of view with diffraction-limited resolution.

Nature Communications, Published online: 11 June 2026; doi:10.1038/s41467-026-74196-5 E-CloudBind improves drug-protein interaction (DPI) prediction by combining electron cloud representations with molecular graphs, enabling robust performance across low-resolution, predicted, and out-of-distribution structures for drug discovery.

A comprehensive computational investigation was performed on the adsorption behaviour of cytarabine anti-cancer drug on the surface of pristine and Fe-decorated (5,5) armchair single-walled carbon nanotube by the density functional theory. The structural parameters, adsorption energies, dipole moments, density of states, the highest occupied molecular orbitals, the lowest unoccupied molecular orb…

America Makes and the National Center for Defense Manufacturing and Machining (NCDMM) have named the recipients of a project call funded by the Office of the Under Secretary of War for Research and Engineering, Manufacturing Technology Office (OSW ManTech), with a combined award of $2M. The initiative, titled Artificial Intelligence for Material Allowables in Additive…