A5050781010- Computational Drug Discovery Methods
- Protein Structure and Dynamics
- Machine Learning in Materials Science
- Bioinformatics and Genomic Networks
- Biomedical Text Mining and Ontologies
- Semantic Web and Ontologies
- Advanced Text Analysis Techniques
- Protein purification and stability
- Asymmetric Hydrogenation and Catalysis
- Chemical Synthesis and Analysis
- Viral Infectious Diseases and Gene Expression in Insects
- Web Data Mining and Analysis
- Carbon dioxide utilization in catalysis
- Microbial Metabolic Engineering and Bioproduction
- CO2 Reduction Techniques and Catalysts
Shanghai Institute of Materia Medica
2023-2024
Chinese Academy of Sciences
2023-2024
ShanghaiTech University
2023-2024
National University of Singapore
2022
Extracting knowledge from complex and diverse chemical texts is a pivotal task for both experimental computational chemists. The still considered to be extremely challenging due the complexity of language scientific literature. This study explored power fine-tuned large models (LLMs) on five intricate text mining tasks: compound entity recognition, reaction role labelling, metal-organic framework (MOF) synthesis information extraction, nuclear magnetic resonance spectroscopy (NMR) data...
Extracting knowledge from complex and diverse chemical texts is a pivotal task for both experimental computational chemists. The still considered to be extremely challenging due the complexity of language scientific literature. This study explored power fine-tuned large models (LLMs) on five intricate text mining tasks: compound entity recognition, reaction role labelling, metal-organic framework (MOF) synthesis information extraction, nuclear magnetic resonance spectroscopy (NMR) data...
Structure-based drug design (SBDD) relies on accurate knowledge of protein structure and ligand-binding conformations. However, most the static conformations obtained by advanced methods such as structural biology de novo folding algorithms often don't meet needs for design. We introduce PackDock, a flexible docking method that combines "conformation selection" "induced fit" mechanisms in two-stage pipeline. The core module this is PackPocket, which uses diffusion model to explore side-chain...
Deep-learning-based classification models are increasingly used for predicting molecular properties in drug development. However, traditional using the Softmax function often give overconfident mispredictions out-of-distribution samples, highlighting a critical lack of accurate uncertainty estimation. Such limitations can result substantial costs and should be avoided during Inspired by advances evidential deep learning Posterior Network, we replaced with normalizing flow to enhance...
ABSTRACT In the field of structure-based drug design, accurately predicting binding conformation ligands to proteins is a long-standing objective. Despite recent advances in deep learning yielding various methods for protein-ligand complex structures, these AI-driven approaches frequently fall short traditional docking practice and often yield structures that lack physical chemical plausibility. To overcome limitations, we present SurfDock, an advanced geometric diffusion network,...
ABSTRACT Developing robust methods for evaluating protein-ligand interactions has been a long-standing problem. Here, we propose novel approach called EquiScore, which utilizes an equivariant heterogeneous graph neural network to integrate physical prior knowledge and characterize in geometric space. To improve generalization performance, constructed dataset PDBscreen designed multiple data augmentation strategies suitable training scoring methods. We also analyzed potential risks of leakage...
<title>Abstract</title> Structure-based drug design (SBDD) relies on accurate knowledge of protein structure and ligand-binding conformations. However, most the static conformations obtained by advanced methods such as structural biology de novo folding algorithms often don’t meet needs for design. We introduce PackDock, a flexible docking method that combines “conformation selection” “induced fit” mechanisms in two-stage pipeline. The core module this is PackPocket, which uses diffusion...
<title>Abstract</title> Understanding protein structure and dynamics is crucial for basic biology drug design. Conventional methods often provide static conformations that inadequately capture flexibility. We present PackDock, a novel approach combining "conformation selection" "induced fit" mechanisms to model protein-ligand interactions. PackDock's core, PackPocket, uses diffusion sample diverse binding pocket or predict ligand-induced changes. validate PackDock through side-chain packing,...
The commonly accepted mechanism of CO2 fixation epoxides to cyclic carbonates catalyzed by multifunctional non-halide organocatalysts is challenged our computational DFT-D3 study, which revealed a new polymerization-like comprising alternate epoxide and activation steps nested pathway. We investigated recently reported coupling with reaction bis-phenolic catalyst. predicted cis/trans product ratio in excellent agreement experimental results. general applicability the supported another...