Solvent-Free Synthesis of Hollow Carbon Nanostructures for Efficient Sodium Storage.

The structural characteristics of hollow carbon nanostructures (HCNs) result in intriguing physicochemical properties and various applications, especially for electrochemical energy storage applications. However, the currently solvent-based template methods to prepare HCNs are still far from meeting the facile, environment-friendly, and scalable demand. Herein, we explored a general and facile solvent-free block copolymer self-assembly approach to prepare various hollow hard carbon nanostructures, including hollow carbon nanofibers, hollow carbon Janus nanotadpoles, hollow carbon spheres, etc. It was found that the obtained HCNs possess abundant active sites, fast pathways for electrons/ions transport, and superior electronic conducting connectivity, which are promising for efficient electrochemical energy storage. Typically, the resultant hollow carbon nanofibers with a thick-walled tube deliver a high reversible capacity (431 mAh g-1) and excellent rate performance (259 mAh g-1 at 800 mA g-1) for sodium ion storage. This intelligent solvent-free block copolymer self-assembly method would inspire the design of hollow hard carbon-based nanostructures for advanced applications in various energy conversion and storage.

Solvent-Free Synthesis of Polymer Spheres and the Activation to Porous Carbon Spheres for Advanced Aluminum-Ion Hybrid Capacitors.

Porous carbon spheres (PCSs) characteristic of perfect symmetry and ideal rheological property have great potential in electrochemical energy storage (EES). However, conventional synthesis of PCSs heavily relies on solution-based methods that may lead to environmental issues. Herein, an environment-friendly solvent-free method toward the facile and mass production of m-phenylenediamine-formaldehyde (MPF) resin spheres, which can be converted into PCSs after carbonization and activation is reported. An ultrahigh productivity of 25.89 g in a 100-mL container and an impressive percent yield of 98.89% can be achieved for the MPF resin spheres, which are further converted into carbon spheres with a reasonable yield of 14.5% after carbonization. When employed as the cathode material for aluminum-ion hybrid capacitors, the obtained PCSs afford a double-layer capacity of ≈200 mAh g-1 , the highest value among reported porous carbon materials for Al-based EES devices. It is anticipated that the solvent-free synthesis method for PCSs developed here may play a significant role in other EES devices, such as magnesium-ion and calcium-ion hybrid capacitors.

Mesoporous Pt@Pt-skin Pt3Ni core-shell framework nanowire electrocatalyst for efficient oxygen reduction.

The design of Pt-based nanoarchitectures with controllable compositions and morphologies is necessary to enhance their electrocatalytic activity. Herein, we report a rational design and synthesis of anisotropic mesoporous Pt@Pt-skin Pt3Ni core-shell framework nanowires for high-efficient electrocatalysis. The catalyst has a uniform core-shell structure with an ultrathin atomic-jagged Pt nanowire core and a mesoporous Pt-skin Pt3Ni framework shell, possessing high electrocatalytic activity, stability and Pt utilisation efficiency. For the oxygen reduction reaction, the anisotropic mesoporous Pt@Pt-skin Pt3Ni core-shell framework nanowires demonstrated exceptional mass and specific activities of 6.69 A/mgpt and 8.42 mA/cm2 (at 0.9 V versus reversible hydrogen electrode), and the catalyst exhibited high stability with negligible activity decay after 50,000 cycles. The mesoporous Pt@Pt-skin Pt3Ni core-shell framework nanowire configuration combines the advantages of three-dimensional open mesopore molecular accessibility and compressive Pt-skin surface strains, which results in more catalytically active sites and weakened chemisorption of oxygenated species, thus boosting its catalytic activity and stability towards electrocatalysis.

Leveraging scaffold information to predict protein-ligand binding affinity with an empirical graph neural network.

Protein-ligand binding affinity prediction is an important task in structural bioinformatics for drug discovery and design. Although various scoring functions (SFs) have been proposed, it remains challenging to accurately evaluate the binding affinity of a protein-ligand complex with the known bound structure because of the potential preference of scoring system. In recent years, deep learning (DL) techniques have been applied to SFs without sophisticated feature engineering. Nevertheless, existing methods cannot model the differential contribution of atoms in various regions of proteins, and the relationship between atom properties and intermolecular distance is also not fully explored. We propose a novel empirical graph neural network for accurate protein-ligand binding affinity prediction (EGNA). Graphs of protein, ligand and their interactions are constructed based on different regions of each bound complex. Proteins and ligands are effectively represented by graph convolutional layers, enabling the EGNA to capture interaction patterns precisely by simulating empirical SFs. The contributions of different factors on binding affinity can thus be transparently investigated. EGNA is compared with the state-of-the-art machine learning-based SFs on two widely used benchmark data sets. The results demonstrate the superiority of EGNA and its good generalization capability.

Electrostatic potential-derived charge: a universal OER performance descriptor for MOFs.

Metal-organic frameworks (MOFs) provide opportunities for the design of high-efficiency catalysts attributed to their high compositional and structural tunability. Meanwhile, the huge number of MOFs poses a great challenge to experimental-intensive development of high-performance functional applications. By taking the computationally feasible and structurally representative trigonal prismatic secondary building units (SBUs) of MOFs as the entry point, we introduce a descriptor-based approach for designing high-performance MOFs for the oxygen evolution reaction (OER). The electrostatic potential-derived charge (ESPC) is identified as a robust and universal OER performance descriptor of MOFs, showing a distinct linear relationship with the onset potentials of OER elemental steps. Importantly, we establish an ESPC-based physical pattern of active site-intermediate binding strength, which interprets the rationality of ESPC as an OER performance descriptor. We further reveal that the SBUs with Ni/Cu as active site atoms while Mn/Fe/Co/Ni as spectator atoms have excellent OER activity through the variation pattern of ESPC along with metal composition. The universal correlation between ESPC and OER activity provides a rational rule for designing high-performance MOF-based OER electrocatalysts and can be easily extended to design functional MOFs for a rich variety of catalytic applications.

An efficient factor for fast screening of high-performance two-dimensional metal-organic frameworks towards catalyzing the oxygen evolution reaction.

Two-dimensional (2D) metal-organic frameworks (MOFs) are promising materials for catalyzing the oxygen evolution reaction (OER) due to their abundant exposed active sites and high specific surface area. However, how to rapidly screen out highly-active 2D MOFs from numerous candidates is still a great challenge. Herein, based on the high-throughput density functional theory (DFT) calculations for 20 kinds of different transition metal-based MOFs, we propose a factor for fast screening of 2D MOFs for the OER under alkaline conditions (pH = 14.0), that is, when the Gibbs free energy change of the O-O bond formation (defined as ΔG 1) is located at ∼1.15 eV, the peak OER performance would be achieved. Based on the high-throughput calculation results, the prediction factor can be further simplified by replacing the Gibbs free energy with the sum of the associated single point energy (SPE) and a binding energy-dependent term. Guided by this factor, we successfully predicted and then obtained the high-performance Ni-based 2D MOFs. This factor would be a practical approach for fast screening of 2D MOF candidates for the OER, and also provide a meaningful reference for the study of other materials.

Fast protein structure comparison through effective representation learning with contrastive graph neural networks.

Protein structure alignment algorithms are often time-consuming, resulting in challenges for large-scale protein structure similarity-based retrieval. There is an urgent need for more efficient structure comparison approaches as the number of protein structures increases rapidly. In this paper, we propose an effective graph-based protein structure representation learning method, GraSR, for fast and accurate structure comparison. In GraSR, a graph is constructed based on the intra-residue distance derived from the tertiary structure. Then, deep graph neural networks (GNNs) with a short-cut connection learn graph representations of the tertiary structures under a contrastive learning framework. To further improve GraSR, a novel dynamic training data partition strategy and length-scaling cosine distance are introduced. We objectively evaluate our method GraSR on SCOPe v2.07 and a new released independent test set from PDB database with a designed comprehensive performance metric. Compared with other state-of-the-art methods, GraSR achieves about 7%-10% improvement on two benchmark datasets. GraSR is also much faster than alignment-based methods. We dig into the model and observe that the superiority of GraSR is mainly brought by the learned discriminative residue-level and global descriptors. The web-server and source code of GraSR are freely available at www.csbio.sjtu.edu.cn/bioinf/GraSR/ for academic use.

Ab-Initio Membrane Protein Amphipathic Helix Structure Prediction Using Deep Neural Networks.

Amphipathic helix (AH)features the segregation of polar and nonpolar residues and plays important roles in many membrane-associated biological processes through interacting with both the lipid and the soluble phases. Although the AH structure has been discovered for a long time, few ab initio machine learning-based prediction models have been reported, due to the limited amount of training data. In this study, we report a new deep learning-based prediction model, which is composed of a residual neural network and the uneven-thresholds decision algorithm. It is constructed on 121 membrane proteins, in total 51640 residue samples, which are curated from an up-to-date membrane protein structure database. Through a rigid 10-fold nested cross-validation experiment, we demonstrate that our model can achieve promising predictions and exceed current state-of-the-art approaches in this field. This presents a new avenue for accurately predicting AHs. Analysis on the contribution of the input residues and some cases further reveals the high interpretability and the generalization of our model.

CoCoPRED: coiled-coil protein structural feature prediction from amino acid sequence using deep neural networks.

MOTIVATION: Coiled-coil is composed of two or more helices that are wound around each other. It widely exists in proteins and has been discovered to play a variety of critical roles in biology processes. Generally, there are three types of structural features in coiled-coil: coiled-coil domain (CCD), oligomeric state and register. However, most of the existing computational tools only focus on one of them. RESULTS: Here, we describe a new deep learning model, CoCoPRED, which is based on convolutional layers, bidirectional long short-term memory, and attention mechanism. It has three networks, i.e. CCD network, oligomeric state network, and register network, corresponding to the three types of structural features in coiled-coil. This means CoCoPRED has the ability of fulfilling comprehensive prediction for coiled-coil proteins. Through the 5-fold cross-validation experiment, we demonstrate that CoCoPRED can achieve better performance than the state-of-the-art models on both CCD prediction and oligomeric state prediction. Further analysis suggests the CCD prediction may be a performance indicator of the oligomeric state prediction in CoCoPRED. The attention heads in CoCoPRED indicate that registers a, b and e are more crucial for the oligomeric state prediction. AVAILABILITY AND IMPLEMENTATION: CoCoPRED is available at http://www.csbio.sjtu.edu.cn/bioinf/CoCoPRED. The datasets used in this research can also be downloaded from the website. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Topology Prediction Improvement of α-helical Transmembrane Proteins Through Helix-tail Modeling and Multiscale Deep Learning Fusion.

Transmembrane proteins (TMPs) play important roles in many biological processes, such as cell recognition and communication. Their structures are crucial for revealing complex functions but are hard to obtain. A variety of computational algorithms have been proposed to fill the gap by predicting structures from primary sequences. In this study, we mainly focus on α-helical TMP and develop a multiscale deep learning pipeline, MemBrain 3.0, to improve topology prediction. This new protocol includes two submodules. The first module is transmembrane helix (TMH) prediction, which features the capability of accurately predicting TMH with the tail part through the incorporation of tail modeling. The prediction engine contains a multiscale deep learning model and a dynamic threshold strategy. The deep learning model is comprised of a small-scale residue-based residual neural network and a large-scale entire-sequence-based residual neural network. Dynamic threshold strategy is designed to binarize the raw prediction scores and solve the under-split problem. The second module is orientation prediction, which consists of a support vector machine (SVM) classifier and a new Max-Min assignment (MMA) strategy. One typical merit of MemBrain 3.0 is the decision mode composed of the dynamic threshold strategy and the MMA strategy, which makes it more effective for hard TMHs, such as half-TMH, back-to-back TMH, and long-TMH. Systematic experiments have demonstrated the efficacy of the new model, which is available at: www.csbio.sjtu.edu.cn/bioinf/MemBrain/.

Assembling a double shell on a diatomite skeleton ternary complex with conductive polypyrrole for the enhancement of supercapacitors.

Double shell FeOOH/PPy on a diatomite ternary complex was assembled via two-step hydrothermal and in situ polymerization routes. The polymerization process generates chemical bonding and introduces oxygen vacancies and mesopores, enhancing the conductivity and electrochemical properties of the electrodes and ensuring structural stability.

Enantioselective photooxygenation of ß-dicarbonyl compounds in batch and flow photomicroreactors.

A series of C-2' modified cinchonine-derived phase-transfer catalysts were synthesized and used in the enantioselective photo-organocatalytic aerobic oxidation of ß-dicarbonyl compounds with excellent yields (up to 97%) and high enantioselectivities (up to 90% ee). Furthermore, the reaction was carried out in a flow photomicroreactor, in which the heterogeneous gas-liquid-liquid asymmetric photocatalytic oxidation reaction was performed affording good yields (up to 97%) and enantioselectivities (up to 86% ee) within 0.89 min.

Monodisperse Carbon Sphere-Constructed Pomegranate-Like Structures for High-Volumetric-Capacitance Supercapacitors.

Porous carbons have been extensively studied in supercapacitors. However, it remains a grand challenge for porous carbons to achieve a volumetric capacitance ( Cv) of over 200 F cm-3 because of the low intrinsic density and limited capacitance. Herein, we propose a pomegranate-like carbon microsphere (PCS) constructed by monodisperse, submicron, N-doped microporous carbon spheres for high-volumetric-capacitance supercapacitors. The assembly of submicron carbon spheres into pomegranate-like structures significantly reduces the required binder amount (2.0 wt %) for electrode preparation, diminishes the interparticle resistance, and most importantly, endows the PCS with a high packing density (0.75 g cm-3). Benefited from the high surface area (1477 m2 g-1), N-doping (3.0 wt %), and high packing density, the PCS demonstrates a high Cv (254 F cm-3), four times that of unassembled monodisperse carbon spheres. This work opens a new avenue to enhance the Cv of porous carbons without compromising the rate capability or cyclability.

Silicon oxides: a promising family of anode materials for lithium-ion batteries.

Silicon oxides have been recognized as a promising family of anode materials for high-energy lithium-ion batteries (LIBs) owing to their abundant reserve, low cost, environmental friendliness, easy synthesis, and high theoretical capacity. However, the extended application of silicon oxides is severely hampered by the intrinsically low conductivity, large volume change, and low initial coulombic efficiency. Significant efforts have been dedicated to tackling these challenges towards practical applications. This Review focuses on the recent advances in the synthesis and lithium storage properties of silicon oxide-based anode materials. To present the progress in a systematic manner, this review is categorized as follows: (i) SiO-based anode materials, (ii) SiO2-based anode materials, (iii) non-stoichiometric SiOx-based anode materials, and (iv) Si-O-C-based anode materials. Finally, future outlook and our personal perspectives on silicon oxide-based anode materials are presented.

The Marriage of the FeN4 Moiety and MXene Boosts Oxygen Reduction Catalysis: Fe 3d Electron Delocalization Matters.

Iron-nitrogen-carbon (Fe-N-C) is hitherto considered as one of the most satisfactory alternatives to platinum for the oxygen reduction reaction (ORR). Major efforts currently are devoted to the identification and maximization of carbon-enclosed FeN4 moieties, which act as catalytically active centers. However, fine-tuning of their intrinsic ORR activity remains a huge challenge. Herein, a twofold activity improvement of pristine Fe-N-C through introducing Ti3 C2 Tx MXene as a support is realized. A series of spectroscopy and magnetic measurements reveal that the marriage of FeN4 moiety and MXene can induce remarkable Fe 3d electron delocalization and spin-state transition of Fe(II) ions. The lower local electron density and higher spin state of the Fe(II) centers greatly favor the Fe d z 2 electron transfer, and lead to an easier oxygen adsorption and reduction on active FeN4 sites, and thus an enhanced ORR activity. The optimized catalyst shows a two- and fivefold higher specific ORR activity than those of pristine catalyst and Pt/C, respectively, even exceeding most Fe-N-C catalysts ever reported. This work opens up a new pathway in the rational design of Fe-N-C catalysts, and reflects the critical influence of Fe 3d electron states in FeN4 moiety supported on MXene in ORR catalysis.

Multiform Sulfur Adsorption Centers and Copper-Terminated Active Sites of Nano-CuS for Efficient Elemental Mercury Capture from Coal Combustion Flue Gas.

Nanostructured copper sulfide synthesized with the assistance of surfactant with nanoscale particle size and high Brunauer-Emmett-Teller surface area was for the first time applied for the capture of elemental mercury (Hg0) from coal combustion flue gas. The optimal operation temperature of nano-CuS for Hg0 adsorption is 75 °C, which indicates that injection of the sorbent between the wet flue gas desulfurization and the wet electrostatic precipitator systems is feasible. This assures that the sorbent is free of the adverse influence of nitrogen oxides. Oxygen (O2) and sulfur dioxide exerted a slight influence on Hg0 adsorption over the nano-CuS. Water vapor was shown to moderately suppress Hg0 capture efficiency via competitive adsorption. The simulated adsorption capacities of nano-CuS for Hg0 under pure nitrogen (N2), N2 + 4% O2, and simulated flue gas reached 122.40, 112.06, and 89.43 mgHg0/g nano-CuS, respectively. Compared to those of traditional commercial activated carbons and metal sulfides, the simulated adsorption capacities of Hg0 over the nano-CuS are at least an order of magnitude higher. Moreover, with only 5 mg loaded in a fixed-bed reactor, the Hg0 adsorption rate reached 11.93-13.56 µg/g min over nano-CuS. This extremely speedy rate makes nano-CuS promising for a future sorbent injection technique. The anisotropic growth of nano-CuS was confirmed by X-ray diffraction analysis and provided a fundamental aspect for nano-CuS surface reconstruction and polysulfide formation. Further X-ray photoelectron spectroscopy and Hg0 temperature-programmed desorption tests showed that the active polysulfide, S-S dimers, and copper-terminated sites contributed primarily to the extremely high Hg0 adsorption capacity and rate. With these advantages, nano-CuS appears to be a highly promising alternative to traditional sorbents for Hg0 capture from coal combustion flue gas.

Fabrication and detection of tissue engineered bone aggregates based on encapsulated human ADSCs within hybrid calcium alginate/bone powder gel-beads in a spinner flask.

Traditional treatment for bone diseases limits their clinical application due to undesirable host immune rejection, limited donator sources and severe pain and suffering for patients. Bone tissue engineering therefore is expected to be a more effective way in treating bone diseases. In the present study, hybrid calcium alginate/bone powder gel-beads with a uniform size distribution, good biocompatibility and osteoinductive capability, were prepared to be used as an in-vitro niche-like matrix. The beads were optimized using 2.5% (w/v) sodium alginate solution, 4.5% (w/v) CaCl2 solution and 5.0mg/mL bone powder using an easy-to-use method. Human ADSCs were cultured and induced into chondrocytes and osteoblasts, respectively. The cells were characterized by histological staining showing the ADSCs were able to maintain their characteristic morphology with multipotent differentiation ability. ADSCs at density of 5 × 10(6)cells/mL were encapsulated into the gel-beads aiming to explore cell expansion under different conditions and the osteogenic induction of ADSCs was verified by specific staining. Results demonstrated that the encapsulated ADSCs expanded 5.6 folds in 10 days under dynamic condition via spinner flask, and were able to differentiate into osteoblasts (OBs) with extensive mineralized nodules forming the bone aggregates over 3 weeks postosteogenic induction. In summary, hybrid gel-beads encapsulating ADSCs are proved to be feasible as a new method to fabricate tissue engineered bone aggregation with potential to treat skeletal injury in the near future.

In vitro culture and harvest of BMMSCs on the surface of a novel thermosensitive glass microcarrier.

Traditional two-dimensional (2D) static culture environment for stem cells followed by enzymatic cell detachment or mechanical treatment is routinely used in research laboratories. However, this method is not ideal as stem cells expand slowly, with cell damage and partial loss of specific stemness. For this reason, a better culture condition is urgently needed to improve stem cell recovery. A novel thermosensitive P(NIPAAm-co-HPM)-g-TMSPM-g-microcarrier was prepared here as a three-dimensional (3D) culture substitute. This novel microcarrier was prepared by grafting NIPAAm and HPM to the surface of glass microcarrier using TMSPM through surface free radical copolymerization. The prepared material was tested in cell culture and via cooling harvest method. We found that NIPAAm was successfully grafted on to the surface of the microcarriers, providing an excellent biocompatible environment for BMMSC adhesion and growth. More importantly, BMMSCs could be fully removed from the thermosensitive glass microcarriers with remained cell viability.