Proteins perform their biological functions through motion. Although high throughput prediction of the three-dimensional static structures of proteins has proved feasible using deep-learning-based methods, predicting the conformational motions remains a challenge. Purely data-driven machine learning methods encounter difficulty for addressing such motions because available laboratory data on conformational motions are still limited. In this work, we develop a method for generating protein allosteric motions by integrating physical energy landscape information into deep-learning-based methods. We show that local energetic frustration, which represents a quantification of the local features of the energy landscape governing protein allosteric dynamics, can be utilized to empower AlphaFold2 (AF2) to predict protein conformational motions. Starting from ground state static structures, this integrative method generates alternative structures as well as pathways of protein conformational motions, using a progressive enhancement of the energetic frustration features in the input multiple sequence alignment sequences. For a model protein adenylate kinase, we show that the generated conformational motions are consistent with available experimental and molecular dynamics simulation data. Applying the method to another two proteins KaiB and ribose-binding protein, which involve large-amplitude conformational changes, can also successfully generate the alternative conformations. We also show how to extract overall features of the AF2 energy landscape topography, which has been considered by many to be black box. Incorporating physical knowledge into deep-learning-based structure prediction algorithms provides a useful strategy to address the challenges of dynamic structure prediction of allosteric proteins.

A new two-dimensional (2D) non-MXene transition metal carbide, Mo3C2, was found using the USPEX code. Comprehensive first-principles calculations show that the Mo3C2 monolayer exhibits thermal, dynamic, and mechanical stability, which can ensure excellent durability in practical applications. The optimized structures of Lix@(3×3)-Mo3C2 (x = 1-36) and Nax@(3×3)-Mo3C2 (x = 1-32) were identified as prospective anode materials. The metallic Mo3C2 sheet exhibits low diffusion barriers of 0.190 eV for Li and 0.118 eV for Na and low average open circuit voltages of 0.31-0.55 V for Li and 0.18-0.48 V for Na. When adsorbing two layers of adatoms, the theoretical energy capacities are 344 and 306 mA h g-1 for Li and Na, respectively, which are comparable to that of commercial graphite. Moreover, the Mo3C2 substrate can maintain structural integrity during the lithiation or sodiation process at high temperature. Considering these features, our proposed Mo3C2 slab is a potential candidate as an anode material for future Li- and Na-ion batteries.

In this chapter, we predict the structure of the Arabidopsis receptor-homology-transmembrane-RING-H2 isoform 1 (RMR1) in complex with the C-terminal sorting determinant of cruciferin (CRU1) by AlphaFold2 using the ColabFold web interface and to perform molecular dynamics simulation to probe the dynamics of the predicted structures. Our results predict that the C-terminal carboxylate group of ctVSD of CRU1 is recognized by the conserved Arg89 of the cargo-binding loop of RMR1 and Arg468 of CRU1 by negative charge residues in the cargo-binding pocket of RMR1. The procedures described here are useful for modeling of other protein complexes.

Spectroscopic data, particularly diffraction data, are essential for materials characterization due to their comprehensive crystallographic information. The current crystallographic phase identification, however, is very time consuming. To address this challenge, we have developed a real-time crystallographic phase identifier based on a convolutional self-attention neural network (CPICANN). Trained on 692 190 simulated powder X-ray diffraction (XRD) patterns from 23 073 distinct inorganic crystallographic information files, CPICANN demonstrates superior phase-identification power. Single-phase identification on simulated XRD patterns yields 98.5 and 87.5% accuracies with and without elemental information, respectively, outperforming JADE software (68.2 and 38.7%, respectively). Bi-phase identification on simulated XRD patterns achieves 84.2 and 51.5% accuracies, respectively. In experimental settings, CPICANN achieves an 80% identification accuracy, surpassing JADE software (61%). Integration of CPICANN into XRD refinement software will significantly advance the cutting-edge technology in XRD materials characterization.

Oriented synthesis of functional materials is a focus of attention in material science. As one of the most important function materials, infrared nonlinear optical materials with large second harmonic generation effects and broad optical band gap are in urgent need. In this work, directed by the theoretical structure prediction, the first series of non-centrosymmetric (NCS) alkali-alkaline earth metal [PS4]-based thiophosphates LiCaPS4 (Ama2), NaCaPS4 (P21), KCaPS4 (Pna21), RbCaPS4 (Pna21), CsCaPS4 (Pna21) were successfully synthesized. Comprehensive characterizations reveal that ACaPS4 could be regarded as promising IR NLO materials, exhibiting wide band gap (3.77-3.86 eV), moderate birefringence (0.027-0.064 at 1064 nm), high laser-induced damage threshold (LIDT, ~10×AGS), and suitable phase-matching second harmonic generation responses (0.4-0.6×AGS). Structure-properties analyses illustrate that the Ca-S bonds show non-ignorable covalent feature, and [PS4] together with [CaSn] units play dominant roles to determine the band gap and SHG response. This work indicates that Li-, Na- and K- analogs may be promising infrared nonlinear optical material candidates, and this is the first successful case of \"prediction to synthesis\" involving infrared (IR) nonlinear optical (NLO) crystals in the thiophosphate system and may provide a new avenue to the design and oriented synthesis of high-performance function materials in the future.

In recent decades, antibodies have emerged as indispensable therapeutics for combating diseases, particularly viral infections. However, their development has been hindered by limited structural information and labor-intensive engineering processes. Fortunately, significant advancements in deep learning methods have facilitated the precise prediction of protein structure and function by leveraging co-evolution information from homologous proteins. Despite these advances, predicting the conformation of antibodies remains challenging due to their unique evolution and the high flexibility of their antigen-binding regions. Here, to address this challenge, we present the Bio-inspired Antibody Language Model (BALM). This model is trained on a vast dataset comprising 336 million 40% nonredundant unlabeled antibody sequences, capturing both unique and conserved properties specific to antibodies. Notably, BALM showcases exceptional performance across four antigen-binding prediction tasks. Moreover, we introduce BALMFold, an end-to-end method derived from BALM, capable of swiftly predicting full atomic antibody structures from individual sequences. Remarkably, BALMFold outperforms those well-established methods like AlphaFold2, IgFold, ESMFold and OmegaFold in the antibody benchmark, demonstrating significant potential to advance innovative engineering and streamline therapeutic antibody development by reducing the need for unnecessary trials. The BALMFold structure prediction server is freely available at https://beamlab-sh.com/models/BALMFold.

Shanxi aged vinegar microbiome encodes a wide variety of bacteriocins. The aim of this study was to mine, screen and characterize novel broad-spectrum bacteriocins from the large-scale microbiome data of Shanxi aged vinegar through machine learning, molecular simulation and activity validation. A total of 158 potential bacteriocins were innovatively mined from 117,552 representative genes based on metatranscriptomic information from the Shanxi aged vinegar microbiome using machine learning techniques and 12 microorganisms were identified to secrete bacteriocins at the genus level. Subsequently, employing AlphaFold2 structure prediction and molecular dynamics simulations, eight bacteriocins with high stability were further screened, and all of them were confirmed to have bacteriostatic activity by the Escherichia coli BL21 expression system. Then, gene_386319 (named LAB-3) and gene_403047 (named LAB-4) with the strongest antibacterial activities were purified by two-step methods and analyzed by mass spectrometry. The two bacteriocins have broad-spectrum antimicrobial activity with minimum inhibitory concentration values of 6.79 μg/mL-15.31 μg/mL against Staphylococcus aureus and Escherichia coli. Furthermore, molecular docking analysis indicated that LAB-3 and LAB-4 could interact with dihydrofolate reductase through hydrogen bonds, salt-bridge forces and hydrophobic forces. These findings suggested that the two bacteriocins could be considered as promising broad-spectrum antimicrobial agents.

X-ray scattering/diffraction tensor tomography techniques are promising methods to acquire the 3D texture information of heterogeneous biological tissues at micrometre resolution. However, the methods suffer from a long overall acquisition time due to multi-dimensional scanning across real and reciprocal space. Here, a new approach is introduced to obtain 3D reciprocal information of each illuminated scanning volume using mathematic modeling, which is equivalent to a physical scanning procedure for collecting the full reciprocal information required for voxel reconstruction. The virtual reciprocal scanning scheme was validated by a simulated 6D wide-angle X-ray diffraction tomography experiment. The theoretical validation of the method represents an important technological advancement for 6D diffraction tensor tomography and a crucial step towards pervasive applications in the characterization of heterogeneous materials.

In recent years, cyclic peptides have emerged as a promising therapeutic modality due to their diverse biological activities. Understanding the structures of these cyclic peptides and their complexes is crucial for unlocking invaluable insights about protein target-cyclic peptide interaction, which can facilitate the development of novel-related drugs. However, conducting experimental observations is time-consuming and expensive. Computer-aided drug design methods are not practical enough in real-world applications. To tackles this challenge, we introduce HighFold, an AlphaFold-derived model in this study. By integrating specific details about the head-to-tail circle and disulfide bridge structures, the HighFold model can accurately predict the structures of cyclic peptides and their complexes. Our model demonstrates superior predictive performance compared to other existing approaches, representing a significant advancement in structure-activity research. The HighFold model is openly accessible at https://github.com/hongliangduan/HighFold.

Oral poisoning can trigger diverse physiological reactions, determined by the toxic substance involved. One such consequence is hyperchloremia, characterized by an elevated level of chloride in the blood and leads to kidney damage and impairing chloride ion regulation. Here, we conducted a comprehensive genome-wide analysis to investigate genes or proteins linked to hyperchloremia. Our analysis included functional enrichment, protein-protein interactions, gene expression, exploration of molecular pathways, and the identification of potential shared genetic factors contributing to the development of hyperchloremia. Functional enrichment analysis revealed that oral poisoning owing hyperchloremia is associated with 4 proteins e.g. Kelch-like protein 3, Serine/threonine-protein kinase WNK4, Serine/threonine-protein kinase WNK1 and Cullin-3. The protein-protein interaction network revealed Cullin-3 as an exceptional protein, displaying a maximum connection of 18 nodes. Insufficient data from transcriptomic analysis indicates that there are lack of information having direct associations between these proteins and human-related functions to oral poisoning, hyperchloremia, or metabolic acidosis. The metabolic pathway of Cullin-3 protein revealed that the derivative is Sulfonamide which play role in, increasing urine output, and metabolic acidosis resulted in hypertension. Based on molecular docking results analysis it found that Cullin-3 proteins has the lowest binding energies score and being suitable proteins. Moreover, no major variations were observed in unbound Cullin-3 and all three peptide bound complexes shows that all systems remain compact during 50 ns simulations. The results of our study revealed Cullin-3 proteins be a strong foundation for the development of potential drug targets or biomarker for future studies.