Articles

Arenes and cyclohexatrienes are pivotal in chemistry, yet synthesizing functionalized variants remains challenging. Here, the authors introduce an iron-catalyzed cyclotrimerization method for internal and terminal heteroatom-substituted alkynes, achieving efficient synthesis of densely functionalized aromatic products.

Synthesis conditions can significantly influence the properties of metal-organic frameworks. Here, the authors report a sustainable and time-controlled synthesis strategy for MIL-125(Ti), demonstrating how reaction duration tunes morphology, pore size, surface area, crystallinity, and biocompatibility, and highlighting synthesis time as a key, eco-friendly parameter for designing high-performance MOFs for biomedical applications.

Water-soluble macrocycles are useful in drug delivery and other applications, yet controlling their host–guest interactions under physiological conditions remains challenging. Here, the authors employ reversible Diels-Alder cycloaddition reactions to control complexation between β-cyclodextrin hosts and anthracene-derived guests in water.

Adverse drug reactions (ADRs) are a severe healthcare concern worldwide, however, accurate prediction of pediatric-specific ADRs remains challenging. Here, the authors present a machine learning approach integrating consensus-driven signal detection and multi-level biological features, significantly enhancing pediatric ADR prediction.

Förster resonance energy transfer (FRET) is essential for studying molecular interactions, yet challenges remain in mapping complex interactions in cellular environments. Here, the authors employ a three-fluorophore FRET-cascade system with time-correlated single photon counting fluorescence lifetime imaging microscopy to elucidate Rap1-interacting adaptor molecule (RIAM)-vinculin interactions in focal adhesions, revealing mechanosensitive roles of vinculin and its interaction with RIAM in a force-independent manner.

Monoclinic zirconia (m-ZrO₂) serves as a crucial barrier against radionuclide release in nuclear fuel Zircaloy cladding, yet the effects of minor actinide incorporation and chemical reactivity remain unclear. Here, the authors reveal that americium enters m-ZrO₂ tetravalently with limited solubility and influences phase transformations, offering insights into behavior of minor actinides in spent nuclear fuel materials.

Achieving molecular chirality with enantiostability in π-conjugated foldamers is challenging due to weak π-π interactions and easy racemization. Here, the authors employ a 3D π–π stacking strategy within a 1,8-diarylnaphthalene framework, achieving exceptional configurational stability and tunable optical properties.

Interstrand crosslinked nucleic acids offer diverse applications in DNA nanotechnology, therapeutics, and biochemistry, yet efficient crosslinking strategies remain underexplored. Here, the authors incorporate two proximal 2′-deoxythioguanosine residues within oligo DNA duplexes for efficient disulfide crosslinking under chemical oxidation and photoinduced reaction.

Despite their intriguing electronic features and promising functional properties, phosphinines remain underexplored because their selective modification is difficult. In this work, the authors demonstrate that halogen–zinc exchanges provide an efficient route to functionalized λ⁵-phosphinines, operating under mild, functional-group-tolerant conditions and enabling access to diverse architectures with notable fluorescence characteristics.

Luminescent lanthanide materials are of value for a range of applications, but most rely on intrashell f–f transitions, limiting their tunability. Here, the authors explore f–d transitions in three-coordinated cerium(III) complexes, achieving orange d–f emissions with photoluminescence quantum yields of up to 62%.

The polymorphic transformation of ritonavir form 1 into less soluble form 2 is a striking example of the risks associated with crystal polymorphism in pharmaceuticals. Here, the authors demonstrate that in silico predictions would have foreseen the existence of ritonavir form 2 and complete the experimental landscape by solving the crystal structure of form 4 by 3D electron diffraction, showing it is unstable and that it exhibits a disordered motif.

Characterizing protein-ligand interactions is crucial for drug development, however, studying weak binders like fragments remains challenging. Here, the authors demonstrate that Hydrogen/Deuterium Exchange Mass Spectrometry (HDX-MS) can effectively characterize fragment binding to Cyclophilin D, offering a valuable tool for fragment-based drug discovery, complementary to other higher resolution techniques.

Traditional nuclear magnetism studies are hindered by low polarization and limited detection methods. Here, the authors utilize the hyperpolarization technique SABRE-SHEATH at 150 nT and 500 nT as well as SQUID sensors to directly detect longitudinal nuclear magnetization to characterize hyperpolarization build-up and spin lattice relaxation in [1-¹³C]-pyruvate, revealing insights into polarization pathways.

Compound–protein interaction prediction is essential for drug discovery, yet current models struggle with accuracy due to reliance on sequence-based data and limited labeled datasets. Here, the authors introduce GenSPARC, a deep learning model utilizing structure-aware protein representations and advanced multimodal molecular representations, achieving generalizability in virtual screening.

Trifluoromethanesulfonyl fluoride (CF₃SO₂F) is a promising alternative to SF₆, a potent greenhouse gas prevalent in electrical grids, but understanding its decomposition pathways and products is crucial for evaluating its environmental feasibility. Here, the authors investigate its thermal decomposition using machine learning-driven molecular dynamics, revealing temperature- and gas pressure-dependent bond-breaking pathways and determining and experimentally validating its decomposition products.

Six-membered metallaromatics have attracted interest, but a general synthetic methodology that enables access to such molecules remains undeveloped. Here, the authors present a skeletal editing approach towards six-membered metallacycles, enabling the creation of osmafluoranthene derivatives with diverse main-group elements.

Electric fields can be used to manipulate molecular spin qubits, but the mechanisms underlying spin-electric coupling are not well understood. Here, the authors investigate the influence of hyperfine coupling between electron and nuclear spins on the mechanism of spin-electric coupling in a 4f molecular qudit.

Traditional AI models in chemistry often struggle with scalability and generalization across tasks. Here, the authors introduce ChemFM, a 3-billion-parameter foundation model pre-trained on 178 million molecules using self-supervised causal language modeling to extract generalizable molecular representations, which outperforms state-of-the-art models in diverse applications in chemical research and antibiotic discovery.

Water molecules are crucial for protein stability and interactions, yet predicting their precise positions around protein structures remain challenging. Here, the authors present SuperWater, a generative AI framework combining a score-based diffusion model and equivariant graph neural networks, to achieve unprecedented accuracy in water placement, demonstrating the capabilities involving protein hydration, protein-ligand binding and protein-protein interactions.

Metabolite alterations are crucial for understanding diseases, yet large-scale untargeted metabolomics faces challenges in signal detection and dataset integration. Here, the authors introduce mzLearn, a data-driven MS¹ signal-detection method that that corrects instrumental drift and enhances signal accuracy and consistency, enabling the development of pre-trained models for untargeted metabolomics.