Yanwei Wang

Developing efficient and robust electrocatalysts for overall water splitting is vital for sustainable green hydrogen production, yet it remains a formidable challenge. In this work, a robust Mo-doped FeNNi3 antiperovskite catalyst (Mo-FeNNi3) is fabricated via a facile two-step strategy involving hydrothermal reaction and subsequent nitrogen treatment. The designed hydrangea-like structure not only endows the electrocatalyst with a robust framework but also exposes abundant available active sites, thereby facilitating efficient mass transport and accelerating reaction kinetics. The as-prepared Mo-FeNNi3 electrocatalyst delivers exceptional catalytic performance, requiring very low overpotentials of 17, 36, and 77 mV to achieve a current density of 10 mA cm–2 for the hydrogen evolution reaction in alkaline, acidic, and neutral media, respectively, alongside an outstanding oxygen evolution reaction overpotential of 223 mV in alkaline solution. Remarkably, when employed in an alkaline electrolyzer (Mo-FeNNi3||Mo-FeNNi3), it delivers a current density of 10 mA cm–2 at a remarkably low cell voltage of 1.472 V, significantly outperforming the commercial RuO2||Pt/C benchmark (1.536 V). The theoretical studies provide evidence that the electronic structure regulation in Mo-FeNNi3 accelerates water dissociation kinetics and optimizes the adsorption/desorption energies of reaction intermediates, thereby dramatically enhancing the hydrogen production efficiency.

Histone H1–mediated phase separation depends on DNA conformational dynamics. Using Holliday junctions we show that H1 accelerates junction isomerization in an ion-dependent manner, linking DNA dynamics to condensate formation and material properties.

This systematic review outlines pivotal advancements in deep learning-driven protein structure prediction and design, focusing on four core models—AlphaFold, RoseTTAFold, RFDiffusion, and ProteinMPNN—developed by 2024 Nobel Laureates in Chemistry: David Baker, Demis Hassabis, and John Jumper. We analyze their technological iterations and collaborative design paradigms, emphasizing breakthroughs in atomic-level structural accuracy, functional protein engineering, and modeling multi-component biomolecular interactions. Key innovations include AlphaFold3's diffusion-based framework for unified biomolecular prediction, RoseTTAFold's three-track architecture integrating sequence and spatial constraints, RFDiffusion's denoising diffusion for de novo protein generation, and ProteinMPNN's inverse folding for sequence–structure co-optimization. Despite major progress in applications such as binder design, nanomaterials, and enzyme engineering, challenges persist in dynamic conformational sampling, multimodal data integration, and generalization to non-canonical targets. We propose future directions, including hybrid physics-AI frameworks and multimodal learning, to bridge gaps between computational design and functional validation in cellular environments.

While the structure and dynamics of spherical active Brownian particles are now relatively well understood, the influence of attractive interactions on these particles continues to present intriguing open questions. In this study, we investigate the structural morphology and particle velocity in two-dimensional systems with specific attractive interactions, informed by purely repulsive systems. This exploration is conducted through computer simulations, with variations in the Péclet number and packing fractions. Our systematic exploration reveals a novel non-monotonic evolution of structural and velocity ordering driven by attraction-activity competition, representing a fundamental departure from conventional active matter behavior. In particular, we identify a critical transition from attraction-dominated to activity-dominated mechanisms as activity intensifies, leading to complex structural reorganization and the emergence of distinct dynamical phases. Moreover, our study reveals significant correlations between topological defects, spatial structures, and particle motion, where defects serve as key mediators of the strong interdependence between structural order and velocity alignment. These findings advance fundamental understanding of non-equilibrium physics and provide novel frameworks for controlling collective behavior in active materials, highlighting the substantial impact of attractive forces in active environments beyond traditional motility-induced phase separation.

Enhancing protein stability through modifications to the N‐ and C‐termini of natural proteins offers the distinct advantages of safety and cost‐effectiveness when compared to the denovo design of proteins. To explore the effect of additional residues at the termini on protein stability, single‐molecule magnetic tweezers are employed to examine the folding and unfolding dynamics of Cold Shock Protein (CSP) with various appended residues (LE‐CSP‐GS, KL‐CSP‐GS, KL‐CSP‐LE). The unfolding rate constant of the LE‐CSP‐GS is an order of magnitude faster than the others, while its folding rate constant decreased by more than an order of magnitude, resulting in upto ≈5 k _B T bigger in folding free energy. Molecular dynamics (MD) simulations revealed that the stability differences are due to additional hydrogen bonds formed by residues K6 and E56. The combination of single‐molecule experiments and MD simulations indicates that additional residues at the termini can significantly affect protein stability and dynamics.

The effect of metal ions on DNA base pairs is crucial for understanding the molecular mechanisms underlying metal-ion-associated gene mutations and has broad applications across various fields. We investigated the interaction between silver ions (Ag⁺) and DNA using single-molecule magnetic tweezers (MT), atomic force microscopy (AFM), and dynamic light scattering (DLS). Our findings reveal that monovalent Ag⁺ can compact DNA directly even at very low concentrations, unlike canonical monovalent ions such as sodium and potassium, which have no effect. We attribute this to the specific binding of Ag⁺ to DNA. For both double-stranded DNA (ds-DNA) and single-stranded DNA (ss-DNA), the critical condensing force (Fc) induced by Ag⁺ initially increases with cationic concentration, reaches a maximum value, and then decreases. We found that the variation in condensing force is due to the rise and fall of silver ions associated with DNA, which is different from the monotonous increase of associated regular cations such as La³⁺ or CoHex³⁺. Notably, the condensing forces for partially denatured DNA by DMSO (including ss-DNA) are larger than those for ds-DNA under the same conditions.

Understanding the intricate interactions governing protein and peptide behavior in liquid–liquid phase separation (LLPS) is crucial for unraveling biological functions and dysfunctions. This study employs a residue-leveled coarse-grained molecular dynamics approach to simulate the phase separation of repetitive polyproline and polyarginine peptides (poly PR) with varying lengths and sequences in solution, considering different concentrations and temperatures. Our findings highlight the crucial role of sequence order in promoting LLPS in peptides with identical lengths of repetitive sequences. Interestingly, repetitive peptides containing fewer than 10 polyarginine repeats exhibit no LLPS, even at salt concentrations up to 3 M. Notably, our simulations align with experimental observations, pinpointing a salt concentration of 2.7 M for PR25-induced LLPS. Utilizing the same methodology, we predict the required salt concentrations for LLPS induction as 1.2 M, 1.5 M, and 2.7 M for PR12, PR15, and PR35, respectively. These predictions demonstrate good agreement with experimental results. Extending our investigation to include the peptide glutamine and arginine (GR15) in DNA solution, our simulations mirror experimental observations of phase separation. To unveil the molecular forces steering peptide phase separation, we introduce a dielectric constant modifier and hydrophobicity disruptor into poly PR systems. Our coarse-grained analysis includes an examination of temperature effects, leading to the inference that both hydrophobic and electrostatic interactions drive phase separation in peptide systems.

The elucidation of the molecular driving forces responsible for Liquid–liquid Phase Separation (LLPS) of proteins and nucleic acids within living cells is crucial for understanding its biological functions and its role in related diseases. In the present study, we investigated the regulation of LLPS in a series of polypeptides with repetitive proline and arginine (PR) sequences by modifying their length and the salt concentration in the solution. Our findings indicate that higher salt concentrations are necessary for LLPS of repetitive PR peptides longer than eight PRs, which emerges as a threshold value. To pinpoint the molecular forces driving the LLPS in peptides, we sequentially introduced various concentrations of hydrophobic disruptors, such as 1,6-hexanediol, and electrostatic regulators, such as ethyl alcohol and 6-Aminocaproic acid. We further modulated the electrostatic interaction by introducing ethyl alcohol and 6-Aminocaproic acid to alter the dielectric constant of the solution. The inclusion of ethyl alcohol intensified the electrostatic interaction between arginine molecules, facilitating LLPS of PR15, while 6-Aminocaproic acid yielded the reverse effect. We deduced that the phase separation in peptide systems is conjointly driven by hydrophobicity and electrostatic interactions. These insights can guide the regulation of LLPS in other peptide and protein systems, and could be pivotal in addressing abnormal aggregations of proteins and nucleic acids.

Liquid–liquid phase separation (LLPS) underlies the formation of membrane-free organelles in eukaryotic cells and plays an important role in the development of some diseases. The phase boundary of metastable liquid–liquid phase separation as well as the cloud point temperature of some globular proteins characterize the phase behavior of proteins and have been widely studied theoretically and experimentally. In the present study, we used a regression and classification neural network to deal with the phase behavior of lysozyme and bovine serum albumin (BSA). We predicted the cloud point temperature and solubility of a lysozyme solution containing sodium chloride by regression and the reentrant phase behavior of BSA in YCl₃ solution containing a surfactant dodecyl dimethyl amine oxide (DDAO) by classification. Specifically, our network model is capable of predicting (a) the solubility of lysozyme in the range: pH 4.0–5.4, temperature 0–25 °C, and NaCl concentration 2–7% (w/v); (b) the cloud point temperature of lysozyme in the range: pH 4.0–4.8, NaCl concentration 2–7%, and lysozyme concentration 0–400 mg/mL; and (c) the phase behavior of BSA in the range: DDAO 1–60 mM, BSA 30–100 mg/mL, and YCl₃ 1–20 mM. We experimentally tested the model at some prediction points with a high accuracy, which means that deep neural networks can be applicable in qualitative and quantitive analysis of liquid–liquid phase separation.