The Journal of Physical Chemistry Letters最新文献

Blood-brain barrier permeability (BBBP) prediction plays a critical role in the drug discovery process, particularly for compounds targeting the central nervous system. While machine learning (ML) has significantly advanced the prediction of BBBP, there remains an urgent need for interpretable ML models that can reveal the physicochemical principles governing BBB permeability. In this study, we propose a multimodal ML framework that integrates molecular fingerprints (Morgan, MACCS, RDK) and image features to improve BBBP prediction. The classification task (BBB-permeable vs nonpermeable) is addressed with a stacking ensemble model combining multiple base classifiers. The proposed framework demonstrates competitive predictive stability, generalization ability, and feature interpretability compared with recent approaches, under comparable evaluation settings. Beyond predictive performance, our framework incorporates Principal Component Analysis (PCA) and Shapley Additive Explanations (SHAP) analysis to highlight key fingerprint features contributing to predictions. The regression task (logBB value prediction) is tackled by a multi-input deep learning framework, incorporating a Transformer encoder for fingerprint processing, a convolutional neural network (CNN) for image feature extraction, and a Multi-Head Attention fusion mechanism to enhance feature interactions. Attention maps derived from the multimodal features reveal token-level relationships within molecular representations. This work provides an interpretable framework for modeling BBBP with enhanced transparency and mechanistic insight and lays the foundation for future studies incorporating transparent descriptors and physics-informed features.

Coccolithophores play critical roles in global carbon and sulfur cycles. They contribute to the carbon cycle through photosynthesis and calcification and the sulfur cycle by producing dimethylsulfoniopropionate (DMSP). Despite their ecological importance, the details and dynamics of methionine metabolism in coccolithophores are poorly understood. Here, we introduce an in situ light-coupled nuclear magnetic resonance (NMR) spectroscopy setup to monitor methionine metabolism directly in coccolithophore cultures under varying environmental conditions. Combining in situ light-coupled NMR spectroscopy and ¹³C magic angle spinning (MAS) spectroscopy, we observed that coccolithophores can take up methionine and convert it into 4-methylthio-2-oxobutyrate (MTOB), which is subsequently secreted into the culture medium, while DMSP was detected only intracellularly. Furthermore, environmental factors, such as elevated temperatures at 24.8 °C, which is 6.8 °C higher than the typical growth temperature for coccolithophores, and darkness, accelerated methionine consumption but reduced its incorporation into proteins and its conversion into MTOB, suggesting a shift toward alternative metabolic pathways under stress. In contrast, seawater acidification had minimal effects on the methionine metabolism. These findings provide new insights into how environmental conditions influence sulfur metabolism in coccolithophores, with potential consequences for their ecological functioning under future climate scenarios.

Surface-enhanced Raman scattering (SERS) technology is highly sensitive but limited by the high cost of noble-metal substrates and the low enhancement of two-dimensional (2D) materials. This work proposes a dual-sided adsorption strategy utilizing warped structures at the cracks and edges of WSe₂ and MoSe₂ nanosheets grown by Chemical Vapor Deposition (CVD). Driven by capillary forces, probe molecules infiltrate the nanosheet-substrate interface, enabling dual-sided adsorption on both the upper and lower surfaces. This approach enhances SERS signals by up to 20-fold, with a detection limit of 10^-10 M, surpassing most reported traditional single-sided adsorption modes. Additionally, it improves stability by isolating the probe molecules from oxygen. This study further enhanced the formation efficiency and coverage area of the dual-sided adsorption mode by leveraging wide warped structures. It offers new perspectives on the application of crack defects and the potential for the development of high-performance and highly stable SERS substrates.

Hybrid polaritonic states can be generated by placing molecules in an optical cavity that is resonant with the energy states of the molecules. There have been investigations of the modification of the chemical kinetics under strong coupling. In this Perspective, we discuss the importance of cavity structure for the modification of chemical dynamics. After reviewing vibrational strong coupling and modification of chemical reactions, we discuss the possibility of using a cavity vacuum field from lower- to higher-order cavity modes by the interaction with multiple complex modes in molecules.

The investigation of cooperative dynamics in H₂O, visible in coherent neutron scattering, has been hindered until now due to the very small signal. Using neutron polarization analysis, we were able, for the first time, to directly measure the coherent neutron scattering signal in light water with unprecedented accuracy. The observed coherent signal is enhanced in the intermediate Q range of 0.2 to 1 Å^-1, providing clear evidence that intermolecular interactions in water extend beyond the distances between nearest neighbors. Our study reveals the existence of a picosecond cooperative process in water, whose nature could be related to the cooperative rearrangements of several water molecules. This process may act as a precursor to large-scale transport related to viscosity. Our results help to improve the understanding of general transport mechanisms at the nanoscale, which can be useful for biomedical technologies or the development of nanofluidic devices.