A5018170478- Computational Drug Discovery Methods
- Machine Learning in Materials Science
- Protein Structure and Dynamics
- Chemical Synthesis and Analysis
- Microbial Natural Products and Biosynthesis
- Machine Learning in Bioinformatics
- AI in cancer detection
- Radiomics and Machine Learning in Medical Imaging
- Catalysis and Oxidation Reactions
- Monoclonal and Polyclonal Antibodies Research
- Medical Imaging Techniques and Applications
- Synthesis and biological activity
- Lung Cancer Diagnosis and Treatment
- Synthesis and Biological Evaluation
- Topic Modeling
- Multimodal Machine Learning Applications
- Advanced Chemical Sensor Technologies
- Optical Wireless Communication Technologies
- COVID-19 diagnosis using AI
- vaccines and immunoinformatics approaches
- PARP inhibition in cancer therapy
- Cancer Cells and Metastasis
- DNA Repair Mechanisms
- Enzyme function and inhibition
- Plant Genetic and Mutation Studies
Korea Research Institute of Chemical Technology
2024
Seoul National University
2011-2024
Korea University
2024
Korea Institute of Science and Technology
2020
Kangwon National University
2020
Korea Institute for Advanced Study
2011
Abstract Designing efficient synthetic routes for a target molecule remains major challenge in organic synthesis. Atom environments are ideal, stand-alone, chemically meaningful building blocks providing high-resolution molecular representation. Our approach mimics chemical reasoning, and predicts reactant candidates by learning the changes of atom associated with reaction. Through careful inspection candidates, we demonstrate as promising descriptors studying reaction route prediction...
Accurate prediction of the binding affinity a protein-ligand complex is essential for efficient and successful rational drug design. Therefore, many methods have been developed. In recent years, since deep learning technology has become powerful, it also implemented to predict affinity. this work, new neural network model that predicts structure Our using ensemble multiple independently trained networks consist channels 3-D convolutional layers. was 3772 complexes from refined set...
A bstract In this preprint, we investigated whether AlphaFold2, AF2, can predict protein-peptide complex structures only with sequence information. We modeled the of 203 complexes from PepBDB DB and 183 PepSet. The were modeling concatenated sequences receptors peptides via poly-glycine linker. found that for more than half test cases, AF2 predicted bound good accuracy, C α -RMSD a peptide < 3.0 Å. For about 40% an accuracy 2.0 Our benchmark results clearly show has great potential to be...
With the rapid improvement of machine translation approaches, neural has started to play an important role in retrosynthesis planning, which finds reasonable synthetic pathways for a target molecule. Previous studies showed that utilizing sequence-to-sequence frameworks is promising approach tackle retrosynthetic planning problem. In this work, we recast problem as language using template-free model. The model trained end-to-end and fully data-driven fashion. Unlike previous models...
Abstract Protein–ligand docking techniques are one of the essential tools for structure‐based drug design. Two major components a successful program an efficient search method and accurate scoring function. In this work, new called LigDockCSA is developed by using powerful global optimization technique, conformational space annealing (CSA), function that combines AutoDock energy piecewise linear potential (PLP) torsion energy. It shown CSA can find lower binding poses than Lamarckian genetic...
Stromal fibrosis in cancer is usually associated with poor prognosis and chemotherapy resistance. It thought to be caused by fibroblasts; however, the exact mechanism not yet well understood. The study aimed identify lineage-specific cancer-associated fibroblast (CAF) subgroup their associations extracellular matrix remodeling clinical significances various tumor types using single-cell bulk RNA sequencing data. Through unsupervised clustering, six subclusters of CAFs were identified,...
Structure-based virtual screening (SBVS) is a crucial computational approach in drug discovery, but its performance sensitive to structural variations. Kinases, which are major targets, exemplify this challenge due active site conformational changes caused by different inhibitor types. Most experimentally determined kinase structures have the DFGin state, potentially biasing SBVS towards type I inhibitors and limiting discovery of diverse scaffolds. We introduce multi-state modeling (MSM)...
Accurate prediction of the binding affinity a protein-ligand complex is essential for efficient and successful rational drug design. In this work, new neural network model that predicts structure developed. Our using ensemble multiple independently trained networks consist channels 3D convolutional layers. was 3740 complexes from refined set PDBbind database tested 270 core set. The benchmark results show correlation coefficient between predicted affinities by our experimental data higher...
This work presents a new template-free neural machine translation method for retrosynthetic reaction prediction by learning the chemical change at substructural level. The proposed effectively solves all issues arising from SMILES-based representation of molecular structures.
Abstract While current computer-aided drug discovery methods offer accuracy or computational efficiency in predicting protein-ligand binding affinities, they face challenges large-scale virtual screenings. Although promising, machine-learning models have shortcomings stemming from limited training sets and docking pose uncertainties. To address these shortcomings, we introduce AK-Score2, a novel interaction prediction model. This model uniquely integrates three independent designed to...
Abstract Structure-based virtual screening (SBVS) is a pivotal computational approach in drug discovery, enabling the identification of potential candidates within vast chemical libraries by predicting their interactions with target proteins. The SBVS relies on receptor protein structures, making it sensitive to structural variations. Kinase, one major targets, known as typical examples an active site conformation change caused type binding inhibitors. Examination human kinase structures...
Abstract Background Interactions between peptide and MHC class II (pMHC-II) are crucial for T-cell recognition immune responses, as MHC-II molecules present fragments to T cells, enabling the distinction self non-self antigens. Accurately predicting pMHC-II binding core is particularly important because it provides insights into interactions receptor engagement. Given high polymorphism peptide-binding promiscuity of molecules, computational prediction methods essential understanding...
Abstract We introduce an advanced model for predicting protein–ligand interactions. Our approach combines the strengths of graph neural networks with physics-based scoring methods. Existing structure-based machine-learning models binding prediction often fall short in practical virtual screening scenarios, hindered by intricacies poses, chemical diversity drug-like molecules, and scarcity crystallographic data complexes. To overcome limitations existing machine learning-based models, we...
Accurate segmentation of lung cancer in pathology slides is a critical step improving patient care. We proposed the ACDC@LungHP (Automatic Cancer Detection and Classification Whole-slide Lung Histopathology) challenge for evaluating different computer-aided diagnosis (CADs) methods on automatic cancer. The 2019 focused (pixel-wise detection) tissue whole slide imaging (WSI), using an annotated dataset 150 training images 50 test from 200 patients. This paper reviews this summarizes top 10...
Accurate prediction of the binding affinity a protein-ligand complex is essential for efficient and successful rational drug design. In this work, new neural network model that predicts structure developed. Our using ensemble multiple independently trained networks consist channels 3D convolutional layers. was 3740 complexes from refined set PDBbind database tested 270 core set. The benchmark results show correlation coefficient between predicted affinities by our experimental data higher...
Microsatellite instability (MSI) is a hypermutable condition caused by DNA mismatch repair system defects, contributing to the development of various cancer types. Recent research has identified Werner syndrome ATP-dependent helicase (WRN) as promising synthetic lethal target for MSI cancers. Herein, we report first discovery thiophen-2-ylmethylene bis-dimedone derivatives novel WRN inhibitors therapy. Initial computational analysis and biological evaluation compound 3aa new scaffold...
This work presents a new template-free neural machine translation method for retrosynthetic reaction prediction by learning the chemical change at substructural level. The proposed effectively solves all issues arising from SMILES-based representation of molecular structures.
<p>This work presents a new template-free neural machine translation method for retrosynthetic reaction prediction by learning the chemical change at substructural level. The proposed effectively solves all issues arising from SMILES-based representation of molecular structures.</p>
Designing efficient synthetic routes for a target molecule remains major challenge in organic synthesis. Atom environments are ideal, stand-alone, chemically meaningful building blocks providing high-resolution molecular representation. Our approach mimics chemical reasoning, and predicts reactant candidates by learning the changes of atom associated with reaction. Through careful inspection candidates, we demonstrate as promising descriptors studying reaction route prediction discovery....
We present a new single-step retrosynthesis prediction method, viz. RetroTRAE, using fragment-based tokenization and the Transformer architecture. RetroTRAE mimics chemical reasoning, predicts reactant candidates by learning changes of atom environments (AEs) associated with reaction. AEs are ideal stand-alone chemically meaningful building blocks providing high-resolution molecular representation. Describing molecule set establishes clear relationship between translated product-reactant...