A5101708276- Protein Structure and Dynamics
- Enzyme Structure and Function
- Lipid Membrane Structure and Behavior
- RNA and protein synthesis mechanisms
- Computational Drug Discovery Methods
- Glycosylation and Glycoproteins Research
- DNA and Nucleic Acid Chemistry
- Machine Learning in Bioinformatics
- Bacteriophages and microbial interactions
- Nanopore and Nanochannel Transport Studies
- Spectroscopy and Quantum Chemical Studies
- Bacterial Genetics and Biotechnology
- Mass Spectrometry Techniques and Applications
- Advanced biosensing and bioanalysis techniques
- Advanced NMR Techniques and Applications
- Chemical Synthesis and Analysis
- Electron Spin Resonance Studies
- Biotin and Related Studies
- Inflammatory mediators and NSAID effects
- Monoclonal and Polyclonal Antibodies Research
- Sulfur Compounds in Biology
- RNA modifications and cancer
- Cellular transport and secretion
- Click Chemistry and Applications
- Lipid metabolism and biosynthesis
Fudan University
2021-2025
Jilin University of Chemical Technology
2024
South China Agricultural University
2024
Capital Medical University
2023-2024
Tsinghua University
2024
Beijing Friendship Hospital
2023-2024
National Clinical Research Center for Digestive Diseases
2024
East China Normal University
2017-2022
New York University Shanghai
2018-2021
New York University
2021
Proper treatment of nonbonded interactions is essential for the accuracy molecular dynamics (MD) simulations, especially in studies lipid bilayers. The use CHARMM36 force field (C36 FF) different MD simulation programs can result disagreements with published simulations performed CHARMM due to differences protocols used treat long-range and 1-4 interactions. In this study, we systematically test C36 FF NAMD, GROMACS, AMBER, OpenMM, CHARMM/OpenMM. A wide range Lennard-Jones (LJ) cutoff...
CHARMM-GUI Membrane Builder, http://www.charmm-gui.org/input/membrane, is a web-based user interface designed to interactively build all-atom protein/membrane or membrane-only systems for molecular dynamics simulations through an automated optimized process. In this work, we describe the new features and major improvements in Builder that allow users robustly realistic biological membrane systems, including (1) addition of lipid types, such as phosphoinositides, cardiolipin (CL),...
Glycolipids (such as glycoglycerolipids, glycosphingolipids, and glycosylphosphatidylinositol) lipoglycans lipopolysaccharides (LPS), lipooligosaccharides (LOS), mycobacterial lipoarabinomannan, mycoplasma lipoglycans) are typically found on the surface of cell membranes play crucial roles in various cellular functions. Characterizing their structure dynamics at molecular level is essential to understand biological roles, but systematic generation glycolipid lipoglycan structures challenging...
Coarse-grained simulations are widely used to study large biological systems. Nonetheless, building such simulation systems becomes nontrivial, especially when membranes with various lipid types involved. Taking advantage of the frameworks in all-atom CHARMM-GUI modules, we have developed Martini Maker for solution, micelle, bilayer, and vesicle as well randomly distributed lipids using force field. supports 82 different flavors field, including polar nonpolar Martini, Dry ElNeDyn (an...
Characterizing glycans and glycoconjugates in the context of three-dimensional structures is important understanding their biological roles developing efficient therapeutic agents. Computational modeling molecular simulation have become an essential tool complementary to experimental methods. Here, we present a computational tool, Glycan Modeler for silico N-/O-glycosylation target protein generation carbohydrate-only systems. In our previous study, developed Reader, web-based detecting...
Abstract N -linked glycosylation is one of the most important, chemically complex and ubiquitous post-translational modifications in all eukaryotes. The -glycans that are covalently linked to proteins involved numerous biological processes. There considerable interest developments general approaches predict structural consequences site-specific understand how these effects can be exploited protein design with advantageous properties. In this study, impacts on structure dynamics...
Accurately predicting changes in protein stability due to mutations is important for engineering and understanding the functional consequences of missense proteins. We have developed DeepDDG, a neural network-based method, use prediction proteins point mutations. The network was trained on more than 5700 manually curated experimental data points able obtain Pearson correlation coefficient 0.48–0.56 three independent test sets, which outperformed 11 other methods. Detailed analysis input...
A complex cell envelope, composed of a mixture lipid types including lipopolysaccharides, protects bacteria from the external environment. Clearly, proteins embedded within various components envelope have an intricate relationship with their local Therefore, to obtain meaningful results, molecular simulations need mimic as far possible this chemically heterogeneous system. However, setting up such systems for computational studies is trivial, and consequently vast majority outer membrane...
Abstract Liquid-liquid phase separation (LLPS) leads to a conversion of homogeneous solution into dense that often resembles liquid droplets, and dilute phase. An increasing number investigations have shown biomolecular condensates formed by LLPS play important roles in both physiology pathology. It has been suggested the behavior proteins would be not only determined sequences, but controlled micro-environmental conditions. Here, we introduce LLPSDB (http://bio-comp.ucas.ac.cn/llpsdb or...
Predicting protein–ligand binding affinity is a central issue in drug design. Various deep learning models have been published recent years, where many of them rely on 3D complex structures as input and tend to focus the single task reproducing affinity. In this study, we developed graph neural network model called PLANET (Protein–Ligand Affinity prediction NETwork). This takes graph-represented structure pocket target protein 2D chemical ligand molecule input. It was trained through...
Computational protein design has a wide variety of applications. Despite its remarkable success, designing for given structure and function is still challenging task. On the other hand, number solved structures rapidly increasing while unique folds reached steady number, suggesting more structural information being accumulated on each fold. Deep learning neural network powerful method to learn such big data set shown superior performance in many machine fields. In this study, we applied deep...
Hydrogen mass repartitioning (HMR) that permits time steps of all-atom molecular dynamics simulation up to 4 fs by increasing the hydrogen atoms has been used in protein and phospholipid bilayers simulations improve conformational sampling. Molecular input generation via CHARMM-GUI now supports HMR for diverse programs. In addition, considering ambiguous pH at bacterial outer membrane surface, different protonation states, either −2e or −1e, phosphate groups lipopolysaccharides (LPS) are...
Human oral bioavailability (HOB) is a key factor in determining the fate of new drugs clinical trials. HOB conventionally measured using expensive and time-consuming experimental tests. The use computational models to evaluate before synthesis will be beneficial drug development process. In this study, total 1588 molecules with data were collected from literature for classifying model that uses consensus predictions five random forest models. shows excellent prediction accuracies on two...
Computational protein design remains a challenging task despite its remarkable success in the past few decades. With rapid progress of deep-learning techniques and accumulation three-dimensional structures, use deep neural networks to learn relationship between sequences structures then automatically sequence for given backbone structure is becoming increasingly feasible. In this study, we developed network named DenseCPD that considers density distribution atoms predicts probability 20...
Coarse-grained (CG) and multiscale simulations are widely used to study large biological systems. However, preparing the simulation system is time-consuming when has multiple components, because each component must be arranged carefully as in protein/micelle or protein/bilayer We have developed CHARMM-GUI PACE CG Builder for building solution, micelle, bilayer systems using force field, a united-atom (UA) model proteins, Martini field water, ions, lipids. The robustness of validated by...
Nanodiscs are discoidal protein–lipid complexes that have wide applications in membrane protein studies. Modeling and simulation of nanodiscs challenging due to the absence structures many scaffold proteins (MSPs) wrap around bilayer. We developed CHARMM‐GUI Nanodisc Builder ( http://www.charmm-gui.org/input/nanodisc ) facilitate setup nanodisc systems by modeling MSPs with defined size known structural features. A total 11 different a diameter from 80 180 Å made available both all‐atom...
The 3C-like proteinase (3CL(pro)) of the severe acute respiratory syndrome (SARS) coronavirus plays a vital role in virus maturation and is proposed to be key target for drug design against SARS. Various vitro studies revealed that only dimer matured 3CL(pro) active. However, as internally encoded gets from replicase polyprotein by autolytic cleavage at both N-terminal C-terminal flanking sites, it unclear whether also needs dimerize first its autocleavage reaction. We constructed large...
Allostery is a common mechanism of controlling many biological processes such as enzyme catalysis, signal transduction, and metabolic regulation. The use allostery to regulate protein activity an important promising strategy in drug discovery network In order modulate by allostery, predictive methods need be developed discover allosteric binding sites. the present study, we new approach identify sites proteins based on coarse-grained two-state Go̅ model. Starting from concept that...