A5025611496- Protein Structure and Dynamics
- Advancements in Battery Materials
- Supercapacitor Materials and Fabrication
- Microgrid Control and Optimization
- Domain Adaptation and Few-Shot Learning
- Graphene research and applications
- RNA and protein synthesis mechanisms
- Advanced Neural Network Applications
- Receptor Mechanisms and Signaling
- Advanced Photocatalysis Techniques
- Power Systems and Renewable Energy
- Optimal Power Flow Distribution
- Multimodal Machine Learning Applications
- Electrocatalysts for Energy Conversion
- Mass Spectrometry Techniques and Applications
- Islanding Detection in Power Systems
- Fuel Cells and Related Materials
- Adsorption and biosorption for pollutant removal
- Heat shock proteins research
- Advanced Nanomaterials in Catalysis
- Advanced battery technologies research
- Advanced Battery Materials and Technologies
- Machine Learning and Data Classification
- Lipid Membrane Structure and Behavior
- Nanomaterials for catalytic reactions
Boston University
2022-2024
Northeastern University
2020-2024
Universidad del Noreste
2023-2024
Tsinghua University
2013-2023
Fudan University Shanghai Cancer Center
2023
Education Department of Hunan Province
2022
Changsha University of Science and Technology
2022
Berkeley College
2021
University of California, Berkeley
2021
Dongbei University of Finance and Economics
2021
Recent work has shown that convolutional networks can be substantially deeper, more accurate, and efficient to train if they contain shorter connections between layers close the input those output. In this paper, we embrace observation introduce Dense Convolutional Network (DenseNet), which connects each layer every other in a feed-forward fashion. Whereas traditional with L have - one its subsequent our network L(L+1)/2 direct connections. For layer, feature-maps of all preceding are used...
Meta-learning has been the most common framework for few-shot learning in recent years. It learns model from collections of classification tasks, which is believed to have a key advantage making training objective consistent with testing objective. However, some works report that by whole-classification, i.e. on whole label-set, it can get comparable or even better embedding than many meta-learning algorithms. The edge between these two lines yet underexplored, and effectiveness remains...
Abstract In this study, soft hydrogel walkers with electro-driven motility for cargo transport have been developed via a facile mould-assisted strategy. The consisting of polyanionic poly(2-acrylamido-2-methylpropanesulfonic acid- co -acrylamide) exhibit an arc looper-like shape two “legs” walking. can reversibly bend and stretch repeated “on/off” electro-triggers in electrolyte solution. Based on such bending/stretching behaviors, the move their to achieve one-directional walking motion...
A fundamental question in protein science is where allosteric hotspots – residues critical for signaling are located, and what properties differentiate them. We carried out deep mutational scanning (DMS) of four homologous bacterial transcription factors (aTFs) to identify built a machine learning model with this data glean the structural molecular hotspots. found be distributed protein-wide rather than being restricted ‘pathways’ linking active sites as commonly assumed. Despite homology,...
For Li2FeSiO4, its P21 space group makes it possibly perfect as a new cathode material for Li-ion batteries (Nishimura et al. J. Am. Chem. Soc. 2008, 130, 13212). this type of Li2MSiO4 (M = Mn, Fe, and Co), the structural, electronic, electrochemical properties have been investigated, using density functional theory with exchange-correlation energy treated generalized gradient approximation (GGA) plus on-site Coulomb correction (+U). Within GGA+U framework, fully lithiated well delithiated...
Abstract Iron–nitrogen–carbon (Fe–N–C) catalysts are considered as the most promising nonprecious metal for oxygen reduction reactions (ORRs). Their synthesis generally involves complex pyrolysis at high temperature, making it difficult to optimize their composition, pore structure, and active sites. This study reports a simple strategy by reacting preformed nitrogen‐doped carbon scaffolds with iron pentacarbonyl, liquid precursor that can effectively form sites nitrogen sites, enabling more...
A reliable nanocasting method has been developed to synthesize mesoporous hybrids of platinum (Pt) nanoparticles decorating tungsten trioxide (WO3). The process began with modification the SBA-15 template carbon polymers and Pt accompanied by adsorption W(6+), which was then converted into m-Pt-WO3 composites heat treatment subsequent removal. synthetic strategy can be easily extended prepare other nanohybrids metal oxide loaded precious composites. Comprehensive characterizations suggest...
New experimental findings continue to challenge our understanding of protein allostery. Recent deep mutational scanning study showed that allosteric hotspots in the tetracycline repressor (TetR) and its homologous transcriptional factors are broadly distributed rather than spanning well-defined structural pathways as often assumed. Moreover, hotspot mutation-induced allostery loss was rescued by additional mutations a degenerate fashion. Here, we develop two-domain thermodynamic model for...
A low-cost and scalable method has been developed to synthesize Fe-decorated N-rich carbon electrocatalysts for the oxygen reduction reaction (ORR) based on pyrolysis of metal carbonyls containing metal-organic frameworks (MOFs). Such a simultaneously optimizes Fe-related active sites porous structure catalysts. Accordingly, best-performing Fe-NC-900-M catalyst shows excellent ORR activity with half-wave potential 0.91 V vs. RHE, exceeding that 40% Pt/C in alkaline media. Furthermore,...
Biomolecular phase separation has emerged as an essential mechanism for cellular organization. How cells respond to environmental stimuli in a robust and sensitive manner build functional condensates at the proper time location is only starting be understood. Recently, lipid membranes have been recognized important regulatory center biomolecular condensation. However, how interplay between behaviors of surface biopolymers may contribute regulation condensation remains elucidated. Using...
Abstract This paper studies in-situ synthesis of Fe 2 O 3 /reduced graphene oxide (rGO) anode materials by different hydrothermal process.Scanning Electron Microscopy (SEM) analysis has found that processes can control the morphology and . The morphologies prepared oleic acid-assisted methods are mainly composed fine spheres, while PVP assists thermal law presents porous ellipsoids. Graphene exhibits typical folds small lumps. X-ray diffraction (XRD) results show is generated in ways. Also,...
Pre-trained language models (e.g., BERT) significantly alleviate two traditional challenging problems for Chinese word segmentation (CWS): ambiguity and out-ofvocabulary (OOV) words.However, such improvements are usually achieved on benchmark datasets not close to an important goal of CWS: practicability (i.e., low complexity as a standalone task high beneficiality downstream tasks).To make trade-off between evaluation CWS, we propose semisupervised neural method via pseudo labels.The...