CONFERENCES
2025
23.
Junyong Park; Sunyong Yoo
Abstract | Links | BibTeX | Tags: Bioinformatics, Generative model, Molecular design
@conference{Park2025b,
title = {Novel Molecular Design via a Scaffold-Aware Transformer with Multi-Scale Attention Mechanisms},
author = {Junyong Park and Sunyong Yoo},
url = {https://bmil.jnu.ac.kr/wp-content/uploads/2025/12/Junyoung-Park-Sunyong-Yoo-Novel-Molecular-Design-via-a-Scaffold-Aware-Transformer-with-Multi-Scale-Attention-Mechanisms.pdf},
year = {2025},
date = {2025-12-17},
publisher = {The 19th International Conference on Data and Text Mining in Biomedical Informatics},
abstract = {Recent advancements in artificial intelligence have demonstrated great potential in accelerating drug discovery by exploring vast chemical spaces and predicting molecular properties. However, conventional molecular generation models have limitations in reflecting desired molecular structures, as they often fail to incorporate specific structural constraints or target properties directly into the generation process. To overcome these limitations, we propose a novel framework that integrates a transformer-based generative model and a graph attention network-based predictive model. The generative model produces molecules with desired structural characteristics by explicitly incorporating scaffold information, while the predictive model estimates the biological activity of the generated molecules. A cyclic learning structure enables the generative and predictive models to interact iteratively, facilitating continuous evaluation and feedback during training. In addition, a multi stage tournament selection with experience memory guides the subsequent training process. Our approach accelerates the identification of scaffold-consistent, high affinity candidates by exploring novel chemical variations around a user-specified scaffold. Experimental results show that the proposed scaffold-aware transformer achieves competitive validity, uniqueness, and novelty, and effectively generates novel compounds with high predicted binding affinity for biological targets. An attention-based analysis extracts atom-level importance scores and highlights the substructures that contribute to the predicted binding affinity, providing interpretable insights into structure-activity relationships. This study provides a practical and interpretable tool for scaffold-conditioned molecular generation.},
keywords = {Bioinformatics, Generative model, Molecular design},
pubstate = {published},
tppubtype = {conference}
}
Recent advancements in artificial intelligence have demonstrated great potential in accelerating drug discovery by exploring vast chemical spaces and predicting molecular properties. However, conventional molecular generation models have limitations in reflecting desired molecular structures, as they often fail to incorporate specific structural constraints or target properties directly into the generation process. To overcome these limitations, we propose a novel framework that integrates a transformer-based generative model and a graph attention network-based predictive model. The generative model produces molecules with desired structural characteristics by explicitly incorporating scaffold information, while the predictive model estimates the biological activity of the generated molecules. A cyclic learning structure enables the generative and predictive models to interact iteratively, facilitating continuous evaluation and feedback during training. In addition, a multi stage tournament selection with experience memory guides the subsequent training process. Our approach accelerates the identification of scaffold-consistent, high affinity candidates by exploring novel chemical variations around a user-specified scaffold. Experimental results show that the proposed scaffold-aware transformer achieves competitive validity, uniqueness, and novelty, and effectively generates novel compounds with high predicted binding affinity for biological targets. An attention-based analysis extracts atom-level importance scores and highlights the substructures that contribute to the predicted binding affinity, providing interpretable insights into structure-activity relationships. This study provides a practical and interpretable tool for scaffold-conditioned molecular generation.
22.
Subhin Seomun; Sunyong Yoo
Abstract | Links | BibTeX | Tags: ADME, Bioinformatics, Deep learning
@conference{Seomun2025,
title = {Cross-species multi-task learning with molecular and ADME descriptors for liver microsomal metabolic stability},
author = {Subhin Seomun and Sunyong Yoo},
url = {https://bmil.jnu.ac.kr/wp-content/uploads/2025/12/Subhin-Seomun-Sunyong-Yoo-Cross-species-multi-task-learning-with-molecular-and-ADME-descriptors-for-liver-microsomal-metabolic-stability.pdf},
year = {2025},
date = {2025-12-17},
urldate = {2025-12-17},
publisher = {The 19th International Conference on Data and Text Mining in Biomedical Informatics},
abstract = {Liver microsomal stability is a key determinant of in vivo compound exposure and efficacy. Although metabolic stability has been extensively studied, linking substructure destabilizing features to absorption, distribution, metabolism, and excretion (ADME) properties remains challenging. Moreover, single-species, single-modality models often generalize poorly. To address these limitations, we propose a cross-species multi-task learning framework that integrates multi-modal molecular representations to predict liver microsomal stability. Specifically, the model leverages three complementary modalities: SMILES-derived fingerprints, molecular graphs, and in silico ADME descriptors. These modalities are learned in a shared network using data from multiple species and subsequently fused via attention mechanisms to form a shared molecular representation, which captures conserved structuremetabolism relationships common across species. Species-specific network capture individual metabolic characteristics and stability predictions for human (HLM), rat (RLM), and mouse liver microsomal (MLM). Under stratified 10-fold cross-validation, mean AUROC was 0.770 Β± 0.001 (HLM), 0.785 Β± 0.001 (RLM), and 0.766 Β± 0.001 (MLM). To understand the chemical basis of metabolic liability, we examined three multi-level perspectives. At the molecular property level, physicochemical descriptors related to enzyme interaction, permeability/transport, and the lipophilicity-polarity axis emerged as dominant predictive drivers. At the substructure level, to pinpoint specific sites of metabolic vulnerability, recurring destabilizing features were identified at alkenes and allylic/benzylic positions, while amide and carbamate carbonyl motifs conferred stability. To elucidate the underlying physicochemical mechanisms, these structural motifs were linked to systematic shifts in logP, solubility, bloodbrain barrier propensity, and efflux liability. Overall, these results indicate that the cross-species integrative model accurately predicts microsomal stability across human, rat, and mouse while providing chemically grounded explanations.},
keywords = {ADME, Bioinformatics, Deep learning},
pubstate = {published},
tppubtype = {conference}
}
Liver microsomal stability is a key determinant of in vivo compound exposure and efficacy. Although metabolic stability has been extensively studied, linking substructure destabilizing features to absorption, distribution, metabolism, and excretion (ADME) properties remains challenging. Moreover, single-species, single-modality models often generalize poorly. To address these limitations, we propose a cross-species multi-task learning framework that integrates multi-modal molecular representations to predict liver microsomal stability. Specifically, the model leverages three complementary modalities: SMILES-derived fingerprints, molecular graphs, and in silico ADME descriptors. These modalities are learned in a shared network using data from multiple species and subsequently fused via attention mechanisms to form a shared molecular representation, which captures conserved structuremetabolism relationships common across species. Species-specific network capture individual metabolic characteristics and stability predictions for human (HLM), rat (RLM), and mouse liver microsomal (MLM). Under stratified 10-fold cross-validation, mean AUROC was 0.770 Β± 0.001 (HLM), 0.785 Β± 0.001 (RLM), and 0.766 Β± 0.001 (MLM). To understand the chemical basis of metabolic liability, we examined three multi-level perspectives. At the molecular property level, physicochemical descriptors related to enzyme interaction, permeability/transport, and the lipophilicity-polarity axis emerged as dominant predictive drivers. At the substructure level, to pinpoint specific sites of metabolic vulnerability, recurring destabilizing features were identified at alkenes and allylic/benzylic positions, while amide and carbamate carbonyl motifs conferred stability. To elucidate the underlying physicochemical mechanisms, these structural motifs were linked to systematic shifts in logP, solubility, bloodbrain barrier propensity, and efflux liability. Overall, these results indicate that the cross-species integrative model accurately predicts microsomal stability across human, rat, and mouse while providing chemically grounded explanations.
21.
Chaewon Kim; Sunyong Yoo
Abstract | Links | BibTeX | Tags: Bioinformatics, Generative model, Transcriptome
@conference{Kim2025,
title = {Predicting Drug-Induced Transcriptional Responses Using Latent Diffusion Model},
author = {Chaewon Kim and Sunyong Yoo},
url = {https://bmil.jnu.ac.kr/wp-content/uploads/2025/12/Chaewon-Kim-Sunyong-Yoo-Predicting-Drug-Induced-Transcriptional-Responses-Using-Latent-Diffusion-Model.pdf},
year = {2025},
date = {2025-12-17},
urldate = {2025-12-17},
publisher = {The 19th International Conference on Data and Text Mining in Biomedical Informatics},
abstract = {Accurate prediction of drug-induced transcriptional responses is essential for drug discovery and precision medicine. Existing computational models, including encoderβdecoder architectures and generative adversarial network-based approaches, achieve reasonable accuracy but often fail to account for geneβgene correlations and generalize to unseen conditions. Here, we present a latent diffusion model that combines a variational autoencoder (VAE) with a diffusion process. The VAE compresses gene expression (GE) profiles into a lowdimensional latent space, where the diffusion process learns the joint probability distribution of latent GE representations and their noisy intermediates. Learning these distributions allow the model to capture geneβgene correlations more effectively. Moreover, our model incorporates multiple perturbation conditionsβincluding cell line, compound, dose, and timeβto enhance generalization performance on unseen conditions. The reverse diffusion process is designed to predict both the mean and variance of the latent representations, which robustly enhances the fidelity of the generated GE profiles. The proposed model demonstrated the highest accuracy in reconstructing perturbed GE profiles compared to previous studies, achieving a root mean squared error (RMSE) of 1.340, a Pearson correlation coefficient of 0.832 and an RΒ² score of 0.669. In addition, the proposed model demonstrated superior performance in preserving geneβgene correlation, as shown by correlation heatmaps, compared to existing approaches. To evaluate the biological relevance of generated transcriptional profiles, we conducted a half-maximal inhibitory concentration prediction task using the generated profiles as model inputs. Our model outperformed the baseline methods, achieving a RMSE of 1.335 and R2 score of 0.819. In conclusion, we demonstrated the potential of diffusion-based generative models as reliable and versatile tools for modeling transcriptional responses, with implications for drug discovery and precision medicine applications.},
keywords = {Bioinformatics, Generative model, Transcriptome},
pubstate = {published},
tppubtype = {conference}
}
Accurate prediction of drug-induced transcriptional responses is essential for drug discovery and precision medicine. Existing computational models, including encoderβdecoder architectures and generative adversarial network-based approaches, achieve reasonable accuracy but often fail to account for geneβgene correlations and generalize to unseen conditions. Here, we present a latent diffusion model that combines a variational autoencoder (VAE) with a diffusion process. The VAE compresses gene expression (GE) profiles into a lowdimensional latent space, where the diffusion process learns the joint probability distribution of latent GE representations and their noisy intermediates. Learning these distributions allow the model to capture geneβgene correlations more effectively. Moreover, our model incorporates multiple perturbation conditionsβincluding cell line, compound, dose, and timeβto enhance generalization performance on unseen conditions. The reverse diffusion process is designed to predict both the mean and variance of the latent representations, which robustly enhances the fidelity of the generated GE profiles. The proposed model demonstrated the highest accuracy in reconstructing perturbed GE profiles compared to previous studies, achieving a root mean squared error (RMSE) of 1.340, a Pearson correlation coefficient of 0.832 and an RΒ² score of 0.669. In addition, the proposed model demonstrated superior performance in preserving geneβgene correlation, as shown by correlation heatmaps, compared to existing approaches. To evaluate the biological relevance of generated transcriptional profiles, we conducted a half-maximal inhibitory concentration prediction task using the generated profiles as model inputs. Our model outperformed the baseline methods, achieving a RMSE of 1.335 and R2 score of 0.819. In conclusion, we demonstrated the potential of diffusion-based generative models as reliable and versatile tools for modeling transcriptional responses, with implications for drug discovery and precision medicine applications.
20.
μ νμ§; μ΄μ¬μΈ; μ μ μ©
Abstract | Links | BibTeX | Tags: Artificial Intelligence, Ethnopharmacology
@conference{nokey,
title = {μ ν΅ μνμμ μ²μ°λ¬Ό λ° νν©λ¬Όμ λ€μ½λ¦¬ν ν¨κ³Ό μλ³ μ°κ΅¬},
author = {μ νμ§ and μ΄μ¬μΈ and μ μ μ©},
url = {https://bmil.jnu.ac.kr/wp-content/uploads/2025/07/μ νμ§-μ ν΅-μνμμ-μ²μ°λ¬Ό-λ°-νν©λ¬Όμ-λ€μ½λ¦¬ν-ν¨κ³Ό-μλ³-μ°κ΅¬.pdf},
year = {2025},
date = {2025-07-04},
urldate = {2025-07-04},
booktitle = {2025 νκ΅λμ§νΈμ½ν
μΈ νν νκ³μ’
ν©νμ λν},
publisher = {νκ΅λμ§νΈμ½ν
μΈ νν},
abstract = {λ³Έ λ
Όλ¬Έμ μ§λ³μ λν μ μ¬μ ν보 μ²μ°λ¬Ό λ° νν©λ¬Όμ μ°κ΄ κ·μΉ λ° κ·Όμ μ± κΈ°λ° λ€νΈμν¬ λΆμμ ν΅ν΄ μλ³ν¨μΌλ‘μ¨ μ ν΅ μνμμμ λ€μ½λ¦¬νμ ν¨κ³Όλ₯Ό λ°νκ³ μ νλ€. μ²μ°λ¬Ό μμ€ λΆμμμ μ λ’°λκ° λμ μ‘°ν©μ μ§λ³μ ν¨κ³Όμ μΌ μ μμΌλ©° νν©λ¬Ό μμ€ λΆμμ μ΄λ₯Ό λ·λ°μΉ¨νλ€.},
keywords = {Artificial Intelligence, Ethnopharmacology},
pubstate = {published},
tppubtype = {conference}
}
λ³Έ λ
Όλ¬Έμ μ§λ³μ λν μ μ¬μ ν보 μ²μ°λ¬Ό λ° νν©λ¬Όμ μ°κ΄ κ·μΉ λ° κ·Όμ μ± κΈ°λ° λ€νΈμν¬ λΆμμ ν΅ν΄ μλ³ν¨μΌλ‘μ¨ μ ν΅ μνμμμ λ€μ½λ¦¬νμ ν¨κ³Όλ₯Ό λ°νκ³ μ νλ€. μ²μ°λ¬Ό μμ€ λΆμμμ μ λ’°λκ° λμ μ‘°ν©μ μ§λ³μ ν¨κ³Όμ μΌ μ μμΌλ©° νν©λ¬Ό μμ€ λΆμμ μ΄λ₯Ό λ·λ°μΉ¨νλ€.
19.
μ‘μ’
μ
; μλ¬ΈμλΉ; μ μ μ©
Abstract | Links | BibTeX | Tags: Deep learning, Interpretability, Transcriptome, Transformer
@conference{nokey,
title = {Transformer κΈ°λ° μλ¬Όνμ κ·Έλν λͺ¨λΈμ νμ©νν΄μ κ°λ₯ν μ½λ¬Ό μ λ μ μ μ λ°ν μμΈ‘},
author = {μ‘μ’
μ
and μλ¬ΈμλΉ and μ μ μ©},
url = {https://bmil.jnu.ac.kr/wp-content/uploads/2025/07/μ‘μ’
μ
-Transformer-κΈ°λ°-μλ¬Όνμ -κ·Έλν-λͺ¨λΈμ-νμ©ν-ν΄μ-κ°λ₯ν-μ½λ¬Ό-μ λ-μ μ μ-λ°ν-μμΈ‘.pdf},
year = {2025},
date = {2025-07-04},
urldate = {2025-07-04},
booktitle = {2025 νκ΅λμ§νΈμ½ν
μΈ νν νκ³μ’
ν©νμ λν},
publisher = {νκ΅λμ§νΈμ½ν
μΈ νν},
abstract = {μ½λ¬Ό μΈν¬ μ©λ μκ°μ λͺ¨λ λ°μν μ½λ¬Ό μ λ μ μ μ λ°ν μμΈ‘μ μ λ°μνκ³Ό λ
μ± νκ°μ νμμ μ΄λ€. κ·Έλ¬λ RNA-seq κΈ°λ° μΈ‘μ μ λΉμ© μκ° λΆλ΄μ΄ ν¬κ³ κΈ°μ‘΄ μ ν κΈ°κ³νμ΅ λͺ¨λΈμ 볡μ‘ν 쑰건 μμ‘΄μ ν¨ν΄μ μΆ©λΆν ν¬μ°©νμ§ λͺ»νλ€. λ³Έ μ°κ΅¬λ μ΄λ₯Ό 극볡νκΈ° μν΄ νν©λ¬Ό SMILES, KEGG κ²½λ‘ κΈ°λ° μΈν¬ κ·Έλν μ©λ μκ° λ²‘ν°λ₯Ό Transformer μΈμ½λλ‘ ν΅ν©ν ν΄μ κ°λ₯ν λ₯λ¬λ λͺ¨λΈμ μ μνλ€. μ μλ λͺ¨λΈμ λλλ§ν¬ μ μ μ λ°νμ λμ μ νλλ‘ μμΈ‘ν λΏ μλλΌ, self-attention λ©μ»€λμ¦μ ν΅ν΄ μ€μν λΆμ νλΆκ΅¬μ‘°μ μ μ μμ κΈ°μ¬λλ₯Ό μλ³νκ³ μκ°νν¨μΌλ‘μ¨ μμΈ‘ κ²°κ³Όμ μλ¬Όνμ ν΄μ κ°λ₯μ±μ ν보νλ€ μ΄λ₯Ό ν΅ν΄ κ³ λΉμ© μ€ν μμ΄λ μ μν ν보 λ¬Όμ§ νμκ³Ό λ
μ± νκ°λ₯Ό κ°μν κ²μΌλ‘ κΈ°λλλ€.},
keywords = {Deep learning, Interpretability, Transcriptome, Transformer},
pubstate = {published},
tppubtype = {conference}
}
μ½λ¬Ό μΈν¬ μ©λ μκ°μ λͺ¨λ λ°μν μ½λ¬Ό μ λ μ μ μ λ°ν μμΈ‘μ μ λ°μνκ³Ό λ
μ± νκ°μ νμμ μ΄λ€. κ·Έλ¬λ RNA-seq κΈ°λ° μΈ‘μ μ λΉμ© μκ° λΆλ΄μ΄ ν¬κ³ κΈ°μ‘΄ μ ν κΈ°κ³νμ΅ λͺ¨λΈμ 볡μ‘ν 쑰건 μμ‘΄μ ν¨ν΄μ μΆ©λΆν ν¬μ°©νμ§ λͺ»νλ€. λ³Έ μ°κ΅¬λ μ΄λ₯Ό 극볡νκΈ° μν΄ νν©λ¬Ό SMILES, KEGG κ²½λ‘ κΈ°λ° μΈν¬ κ·Έλν μ©λ μκ° λ²‘ν°λ₯Ό Transformer μΈμ½λλ‘ ν΅ν©ν ν΄μ κ°λ₯ν λ₯λ¬λ λͺ¨λΈμ μ μνλ€. μ μλ λͺ¨λΈμ λλλ§ν¬ μ μ μ λ°νμ λμ μ νλλ‘ μμΈ‘ν λΏ μλλΌ, self-attention λ©μ»€λμ¦μ ν΅ν΄ μ€μν λΆμ νλΆκ΅¬μ‘°μ μ μ μμ κΈ°μ¬λλ₯Ό μλ³νκ³ μκ°νν¨μΌλ‘μ¨ μμΈ‘ κ²°κ³Όμ μλ¬Όνμ ν΄μ κ°λ₯μ±μ ν보νλ€ μ΄λ₯Ό ν΅ν΄ κ³ λΉμ© μ€ν μμ΄λ μ μν ν보 λ¬Όμ§ νμκ³Ό λ
μ± νκ°λ₯Ό κ°μν κ²μΌλ‘ κΈ°λλλ€.
18.
λμλΉ; μ μ μ°; μ΅ν¬μ; μ μ μ©
Abstract | Links | BibTeX | Tags: CYP450, Deep learning
@conference{nokey,
title = {λ₯λ¬λ κΈ°λ° Cytochrome P450 2D6 μ μ μ λ€νμ±κ³Ό μ½λ¬Ό νΉμ΄μ λμ¬ κΈ°λ₯ ννν μμΈ‘ μ°κ΅¬},
author = {λμλΉ and μ μ μ° and μ΅ν¬μ and μ μ μ©},
url = {https://bmil.jnu.ac.kr/wp-content/uploads/2025/07/λμλΉ-λ₯λ¬λ-κΈ°λ°-Cytochrome-P450-2D6-μ μ μ-λ€νμ±κ³Ό-μ½λ¬Ό-νΉμ΄μ -λμ¬-κΈ°λ₯-ννν-μμΈ‘-μ°κ΅¬.pdf},
year = {2025},
date = {2025-07-04},
urldate = {2025-07-04},
booktitle = {2025 νκ΅λμ§νΈμ½ν
μΈ νν νκ³μ’
ν©νμ λν},
publisher = {νκ΅λμ§νΈμ½ν
μΈ νν},
abstract = {CYP2D6λ μμμμ μ¬μ©λλ μ½λ¬Όμ 25%λ₯Ό λμ¬νλ€. λμ λ€νμ±μ νΉμ§μΌλ‘ νλ CYP2D6μ μ μ μ λ³μ΄λ μ½λ¬Ό λμ¬μμ κ°μΈ κ° ν° μ°¨μ΄λ₯Ό μ΄λν μ μμ΄λ©° μ΄λ μΉλ£ λ°μμ μ°¨μ΄μ λΆμμ©μΌλ‘ μ΄μ΄μ§ μ μλ€. κΈ°μ‘΄μ CYP2D6 μ½λ¬Ό λμ¬ ννν λΆλ₯ λ°©μμ μ μν΄ λμ¬λλ λͺ κ°μ§ μ½λ¬Όμ μμκ²°κ³Όλ₯Ό λ°νμΌλ‘ λ³μ΄μ²΄μ μ μλ₯Ό λ§€κΈ°κ³ μ΄λ₯Ό ν΅ν΄ λͺ¨λ μ½λ¬Όμ λμ¬ λ₯λ ₯μ μμΈ‘νλ λ°©λ²μ΄μλ€ νμ§λ§ μ΄ λ°©λ²μ μ½λ¬Όλ§λ€ λ€λ₯Έ νΉμ±μ λ°μνμ§ λͺ»νκΈ°μ λͺ¨λ μ½λ¬Όμ μΌκ΄μ μΌλ‘ μ μ©νκΈ°μλ νκ³κ° μλ€. λ°λΌμ λ³Έ
μ°κ΅¬μμλ CYP2D6λ³μ΄μ²΄μ μ½λ¬Όμ λν μμκ²°κ³Όλ₯Ό μ§μ νμ©νμ¬ λ°μ΄ν° λΌλ²¨λ§μ μννκ³ λ₯λ¬λμ νμ©ν μ½λ¬Ό λμ¬ ννν μμΈ‘ λͺ¨λΈμ κ°λ°νμλ€},
keywords = {CYP450, Deep learning},
pubstate = {published},
tppubtype = {conference}
}
CYP2D6λ μμμμ μ¬μ©λλ μ½λ¬Όμ 25%λ₯Ό λμ¬νλ€. λμ λ€νμ±μ νΉμ§μΌλ‘ νλ CYP2D6μ μ μ μ λ³μ΄λ μ½λ¬Ό λμ¬μμ κ°μΈ κ° ν° μ°¨μ΄λ₯Ό μ΄λν μ μμ΄λ©° μ΄λ μΉλ£ λ°μμ μ°¨μ΄μ λΆμμ©μΌλ‘ μ΄μ΄μ§ μ μλ€. κΈ°μ‘΄μ CYP2D6 μ½λ¬Ό λμ¬ ννν λΆλ₯ λ°©μμ μ μν΄ λμ¬λλ λͺ κ°μ§ μ½λ¬Όμ μμκ²°κ³Όλ₯Ό λ°νμΌλ‘ λ³μ΄μ²΄μ μ μλ₯Ό λ§€κΈ°κ³ μ΄λ₯Ό ν΅ν΄ λͺ¨λ μ½λ¬Όμ λμ¬ λ₯λ ₯μ μμΈ‘νλ λ°©λ²μ΄μλ€ νμ§λ§ μ΄ λ°©λ²μ μ½λ¬Όλ§λ€ λ€λ₯Έ νΉμ±μ λ°μνμ§ λͺ»νκΈ°μ λͺ¨λ μ½λ¬Όμ μΌκ΄μ μΌλ‘ μ μ©νκΈ°μλ νκ³κ° μλ€. λ°λΌμ λ³Έ
μ°κ΅¬μμλ CYP2D6λ³μ΄μ²΄μ μ½λ¬Όμ λν μμκ²°κ³Όλ₯Ό μ§μ νμ©νμ¬ λ°μ΄ν° λΌλ²¨λ§μ μννκ³ λ₯λ¬λμ νμ©ν μ½λ¬Ό λμ¬ ννν μμΈ‘ λͺ¨λΈμ κ°λ°νμλ€
μ°κ΅¬μμλ CYP2D6λ³μ΄μ²΄μ μ½λ¬Όμ λν μμκ²°κ³Όλ₯Ό μ§μ νμ©νμ¬ λ°μ΄ν° λΌλ²¨λ§μ μννκ³ λ₯λ¬λμ νμ©ν μ½λ¬Ό λμ¬ ννν μμΈ‘ λͺ¨λΈμ κ°λ°νμλ€
17.
κΉμ±μ; μ λͺ
ν; κΉλ―Όκ±΄; μ μ μ©
Abstract | Links | BibTeX | Tags: Artificial Intelligence, Bioinformatics, Drugs, Transcriptome
@conference{nokey,
title = {Conditional Diffusion Model κΈ°λ° μ½λ¬Όλ‘ μΈν μ μ¬μ²΄ λ°μ μμΈ‘},
author = {κΉμ±μ and μ λͺ
ν and κΉλ―Όκ±΄ and μ μ μ©},
url = {https://bmil.jnu.ac.kr/wp-content/uploads/2025/07/κΉμ±μ-Conditional-Diffusion-Model-κΈ°λ°-μ½λ¬Όλ‘-μΈν-μ μ¬μ²΄-λ°μ-μμΈ‘.pdf},
year = {2025},
date = {2025-07-04},
urldate = {2025-07-04},
booktitle = {2025 νκ΅λμ§νΈμ½ν
μΈ νν νκ³μ’
ν©νμ λν},
publisher = {νκ΅λμ§νΈμ½ν
μΈ νν},
abstract = {λ³Έ λ
Όλ¬Έμμλ Conditional Diffusion Model κΈ°λ° κ΅λ 쑰건μ κ³ λ €ν μ μ¬μ²΄ λ³ν μμΈ‘ μ¬μΈ΅ μμ± λͺ¨λΈμ μκ°νλ€ μ²λ¦¬ν νν©λ¬Ό μ 보μ λλΆμ΄ μ²λ¦¬μ©λκ³Ό μκ° μΈν¬μ£Όμ κΈ°μ μ μ μ λ°ν μ 보λ₯Ό μ¬μ©ν¨μΌλ‘μ¨ μ λ°ν μ μ¬μ²΄ λ³ν μμΈ‘μ κ°λ₯νκ² νλ€ λ°λΌμ λ³Έ λͺ¨λΈμ΄ μμ±ν μ μ¬μ²΄ λ³ν λ°μ΄ν°λ₯Ό νμ©ν¨μΌλ‘μ¨ μ½λ¬Όμ λν μ΄ν΄λλ₯Ό ν₯μνκ³ μ μ½ κ°λ° λ° μ λ° μλ£ κΈ°μ μ λ°μ λ±μ κΈ°μ¬ν μ μλ κ°λ₯μ±μ 보μ¬μ€λ€.},
keywords = {Artificial Intelligence, Bioinformatics, Drugs, Transcriptome},
pubstate = {published},
tppubtype = {conference}
}
λ³Έ λ
Όλ¬Έμμλ Conditional Diffusion Model κΈ°λ° κ΅λ 쑰건μ κ³ λ €ν μ μ¬μ²΄ λ³ν μμΈ‘ μ¬μΈ΅ μμ± λͺ¨λΈμ μκ°νλ€ μ²λ¦¬ν νν©λ¬Ό μ 보μ λλΆμ΄ μ²λ¦¬μ©λκ³Ό μκ° μΈν¬μ£Όμ κΈ°μ μ μ μ λ°ν μ 보λ₯Ό μ¬μ©ν¨μΌλ‘μ¨ μ λ°ν μ μ¬μ²΄ λ³ν μμΈ‘μ κ°λ₯νκ² νλ€ λ°λΌμ λ³Έ λͺ¨λΈμ΄ μμ±ν μ μ¬μ²΄ λ³ν λ°μ΄ν°λ₯Ό νμ©ν¨μΌλ‘μ¨ μ½λ¬Όμ λν μ΄ν΄λλ₯Ό ν₯μνκ³ μ μ½ κ°λ° λ° μ λ° μλ£ κΈ°μ μ λ°μ λ±μ κΈ°μ¬ν μ μλ κ°λ₯μ±μ 보μ¬μ€λ€.
16.
κΉμλ―Ό; μ΄λν; μ μ μ©
Abstract | Links | BibTeX | Tags: Graph attention network, Interpretability, Transformer
@conference{κΉμλ―Ό2025,
title = {κ·Έλν νΈλμ€ν¬λ¨Έλ₯Ό μ΄μ©ν νμμ μ‘°ν©μ μλμ§ ν¨κ³Ό μμΈ‘},
author = {κΉμλ―Ό and μ΄λν and μ μ μ©},
url = {https://bmil.jnu.ac.kr/wp-content/uploads/2025/07/κΉμλ―Ό-κ·Έλν-νΈλμ€ν¬λ¨Έλ₯Ό-μ΄μ©ν-νμμ -μ‘°ν©μ-μλμ§-ν¨κ³Ό-μμΈ‘.pdf},
year = {2025},
date = {2025-07-04},
urldate = {2025-07-04},
booktitle = {2025 νκ΅λμ§νΈμ½ν
μΈ νν νκ³μ’
ν©νμ λν},
publisher = {νκ΅λμ§νΈμ½ν
μΈ νν},
abstract = {μ½λ¬Ό μ‘°ν© μΉλ£λ μ μΉλ£μ μμ΄ μ λ§ν μΉλ£ μ λ΅μΌλ‘ λ μ€λ₯΄κ³ μλ€ κ·Έλ¬λ μ½λ¬Όμ μκ° μ¦κ°ν¨μ λ°λΌ ν¨κ³Όμ μΈ μ½λ¬Ό μ‘°ν©μ μλ³νλ κ²μ μ¬μ ν μ΄λ €μ΄ κ³Όμ μ΄λ€ κΈ°μ‘΄ μ°κ΅¬λ€μ λΆμ κ·Έλνμ ꡬ쑰μ νΉμ§μ μΆ©λΆνμΏ λ°μνμ§ λͺ»νκ³ μλμ§ ν¨κ³Όμ μ€μν μ μ μμ λν λΆμμ΄ λΆμ‘±νλ€λ νκ³κ° μ‘΄μ¬νλ€ λ³Έ λ
Όλ¬Έμμλ μ΄λ₯Ό ν΄κ²°νκΈ° μν΄ κ·Έλν νΈλμ€ν¬λ¨Έμ
κ²μ΄ν
λ©μ»€λμ¦μ κ²°ν©ν λͺ¨λΈμ μ μνλ€ μ μλ λͺ¨λΈμ κΈ°μ‘΄ λ°©λ²λ€ λ³΄λ€ μ°μν μ±λ₯μ 보μκ³ κ²μ΄ν
λ©μ»€λμ¦μ ν΅ν΄ μλμ§ ν¨κ³Όμ μ€μν μ μ μλ€μ μλ³ν¨μΌλ‘μ¨ ν΄μ κ°λ₯μ±μ ν보νμλ€ μ΄λ₯Ό ν΅ν΄ μ½λ¬Ό μ‘°ν© μλ³μ μν μ λ§ν λκ΅¬λ‘ νμ©λ μ μμ κ²μΌλ‘ κΈ°λλλ€.},
keywords = {Graph attention network, Interpretability, Transformer},
pubstate = {published},
tppubtype = {conference}
}
μ½λ¬Ό μ‘°ν© μΉλ£λ μ μΉλ£μ μμ΄ μ λ§ν μΉλ£ μ λ΅μΌλ‘ λ μ€λ₯΄κ³ μλ€ κ·Έλ¬λ μ½λ¬Όμ μκ° μ¦κ°ν¨μ λ°λΌ ν¨κ³Όμ μΈ μ½λ¬Ό μ‘°ν©μ μλ³νλ κ²μ μ¬μ ν μ΄λ €μ΄ κ³Όμ μ΄λ€ κΈ°μ‘΄ μ°κ΅¬λ€μ λΆμ κ·Έλνμ ꡬ쑰μ νΉμ§μ μΆ©λΆνμΏ λ°μνμ§ λͺ»νκ³ μλμ§ ν¨κ³Όμ μ€μν μ μ μμ λν λΆμμ΄ λΆμ‘±νλ€λ νκ³κ° μ‘΄μ¬νλ€ λ³Έ λ
Όλ¬Έμμλ μ΄λ₯Ό ν΄κ²°νκΈ° μν΄ κ·Έλν νΈλμ€ν¬λ¨Έμ
κ²μ΄ν λ©μ»€λμ¦μ κ²°ν©ν λͺ¨λΈμ μ μνλ€ μ μλ λͺ¨λΈμ κΈ°μ‘΄ λ°©λ²λ€ λ³΄λ€ μ°μν μ±λ₯μ 보μκ³ κ²μ΄ν λ©μ»€λμ¦μ ν΅ν΄ μλμ§ ν¨κ³Όμ μ€μν μ μ μλ€μ μλ³ν¨μΌλ‘μ¨ ν΄μ κ°λ₯μ±μ ν보νμλ€ μ΄λ₯Ό ν΅ν΄ μ½λ¬Ό μ‘°ν© μλ³μ μν μ λ§ν λκ΅¬λ‘ νμ©λ μ μμ κ²μΌλ‘ κΈ°λλλ€.
κ²μ΄ν λ©μ»€λμ¦μ κ²°ν©ν λͺ¨λΈμ μ μνλ€ μ μλ λͺ¨λΈμ κΈ°μ‘΄ λ°©λ²λ€ λ³΄λ€ μ°μν μ±λ₯μ 보μκ³ κ²μ΄ν λ©μ»€λμ¦μ ν΅ν΄ μλμ§ ν¨κ³Όμ μ€μν μ μ μλ€μ μλ³ν¨μΌλ‘μ¨ ν΄μ κ°λ₯μ±μ ν보νμλ€ μ΄λ₯Ό ν΅ν΄ μ½λ¬Ό μ‘°ν© μλ³μ μν μ λ§ν λκ΅¬λ‘ νμ©λ μ μμ κ²μΌλ‘ κΈ°λλλ€.
15.
κ°λ―ΌκΈ°; μ‘μ€μ£Ό; μ μ μ©
Abstract | Links | BibTeX | Tags: Artificial Intelligence, Knowledge graph
@conference{κ°λ―ΌκΈ°2025,
title = {μ§μ κ·Έλν μλ² λ© κΈ°λ° μ½λ¬Ό-μν μνΈμμ© μμΈ‘ μ°κ΅¬},
author = {κ°λ―ΌκΈ° and μ‘μ€μ£Ό and μ μ μ©},
url = {https://bmil.jnu.ac.kr/wp-content/uploads/2025/07/κ°λ―ΌκΈ°-μ§μ-κ·Έλν-μλ² λ©-κΈ°λ°-μ½λ¬Ό-μν-μνΈμμ©-μμΈ‘-μ°κ΅¬-1.pdf},
year = {2025},
date = {2025-07-04},
urldate = {2025-07-04},
booktitle = {2025 νκ΅λμ§νΈμ½ν
μΈ νν νκ³μ’
ν©νμ λν},
publisher = {νκ΅λμ§νΈμ½ν
μΈ νν},
abstract = {μν μ½λ¬Ό μνΈμμ© μ νμ μμ μ μ€μν μν μμμ΄μ§λ§ κΈ°μ‘΄ μμΈ‘ λ°©λ²λ€μ 볡μ‘ν μννμ κ΄κ³λ₯Ό μΆ©λΆν κ³ λ €νμ§ λͺ»νλ€ λ³Έ λ
Όλ¬Έμμλ μ§μ κ·Έλν μ κ²½λ§κ³Ό cross-attention λ©μ»€λμ¦μ κ²°ν©νμ¬ μ½λ¬Όλ³ λ§₯λ½μμ κ΄λ ¨μ± λμ μν νΉμ±μ κ°μ‘°ν¨μΌλ‘μ¨ FDIλ₯Ό μμΈ‘νλ λͺ¨λΈμ μ μνλ€ λ€μ€ μμν λ°μ΄ν°λ² μ΄μ€λ₯Ό ν΅ν©ν μ§μ κ·Έλν κΈ°λ°μΌλ‘ μνμ 볡ν©μ μνν ν¨κ³Όλ₯Ό λͺ¨λΈλ§ν κ²°κ³Ό κΈ°μ‘΄ λ°©λ²λ€ λλΉ μ°μν μμΈ‘ μ±λ₯μ λ¬μ±νμ¬ μ
μ νκ²½μμμ FDI μν κ΄λ¦¬μ κΈ°μ¬ν μ μμ κ²μΌλ‘ κΈ°λλλ€.},
keywords = {Artificial Intelligence, Knowledge graph},
pubstate = {published},
tppubtype = {conference}
}
μν μ½λ¬Ό μνΈμμ© μ νμ μμ μ μ€μν μν μμμ΄μ§λ§ κΈ°μ‘΄ μμΈ‘ λ°©λ²λ€μ 볡μ‘ν μννμ κ΄κ³λ₯Ό μΆ©λΆν κ³ λ €νμ§ λͺ»νλ€ λ³Έ λ
Όλ¬Έμμλ μ§μ κ·Έλν μ κ²½λ§κ³Ό cross-attention λ©μ»€λμ¦μ κ²°ν©νμ¬ μ½λ¬Όλ³ λ§₯λ½μμ κ΄λ ¨μ± λμ μν νΉμ±μ κ°μ‘°ν¨μΌλ‘μ¨ FDIλ₯Ό μμΈ‘νλ λͺ¨λΈμ μ μνλ€ λ€μ€ μμν λ°μ΄ν°λ² μ΄μ€λ₯Ό ν΅ν©ν μ§μ κ·Έλν κΈ°λ°μΌλ‘ μνμ 볡ν©μ μνν ν¨κ³Όλ₯Ό λͺ¨λΈλ§ν κ²°κ³Ό κΈ°μ‘΄ λ°©λ²λ€ λλΉ μ°μν μμΈ‘ μ±λ₯μ λ¬μ±νμ¬ μ
μ νκ²½μμμ FDI μν κ΄λ¦¬μ κΈ°μ¬ν μ μμ κ²μΌλ‘ κΈ°λλλ€.
μ νκ²½μμμ FDI μν κ΄λ¦¬μ κΈ°μ¬ν μ μμ κ²μΌλ‘ κΈ°λλλ€.
2024
14.
Yeabean Na; Junho Kim; Myung-Gyun Kang; Sunyong Yoo
Abstract | Links | BibTeX | Tags: Bioinformatics, Deep learning, Drugs
@conference{Yoo2024,
title = {A Multimodal Deep Learning Approach for Predicting Drug Metabolism According to the CYP2D6 Genetic Variation},
author = {Yeabean Na and Junho Kim and Myung-Gyun Kang and Sunyong Yoo},
url = {https://dtmbio.net/},
year = {2024},
date = {2024-01-02},
urldate = {2024-01-02},
publisher = {The 18th International Conference on Data and Text Mining in Biomedical Informatics},
abstract = {Background Cytochrome P450 2D6 (CYP2D6) is involved in metabolizing up to 25% of the drugs commonly used in clinics. Characterized by high polymorphisms, CYP2D6 is one of the key pharmacogenes in pharmacogenomics. This genetic variability can lead to significant inter-patient differences in drug metabolism, resulting in differential therapeutic responses and adverse effects. However, conducting in vivo or in vitro experiments for each CYP2D6 variant across various drugs is time-consuming, ethically challenging, and expensive. Given these constraints, In silico modeling approaches for predicting the drug metabolism profiles of CYP2D6 variants are a critical necessity.
Methods A multimodal deep learning approach that combined CYP2D6 genotype data and drug structural information was used in this study. A Convolutional Neural Network (CNN) was used to encode the genotype data, and a Graph Convolutional Network (GCN) was used to decode the drug structures. These diverse data types were then integrated into a multimodal model to predict drug metabolism.
Results A comparative analysis was conducted between a CNN model utilizing solely the CYP2D6 genotype data and a multimodal model incorporating both genotype and drug-specific information. The multimodal approach demonstrated better performance across all evaluated metrics. An additional experiment predicting drug metabolism on unseen drug data also performed well.
Conclusions This model is anticipated to enhance the prediction of metabolic capacity in previously uncharacterized CYP2D6 variants, potentially reducing adverse drug reactions.},
keywords = {Bioinformatics, Deep learning, Drugs},
pubstate = {published},
tppubtype = {conference}
}
Background Cytochrome P450 2D6 (CYP2D6) is involved in metabolizing up to 25% of the drugs commonly used in clinics. Characterized by high polymorphisms, CYP2D6 is one of the key pharmacogenes in pharmacogenomics. This genetic variability can lead to significant inter-patient differences in drug metabolism, resulting in differential therapeutic responses and adverse effects. However, conducting in vivo or in vitro experiments for each CYP2D6 variant across various drugs is time-consuming, ethically challenging, and expensive. Given these constraints, In silico modeling approaches for predicting the drug metabolism profiles of CYP2D6 variants are a critical necessity.
Methods A multimodal deep learning approach that combined CYP2D6 genotype data and drug structural information was used in this study. A Convolutional Neural Network (CNN) was used to encode the genotype data, and a Graph Convolutional Network (GCN) was used to decode the drug structures. These diverse data types were then integrated into a multimodal model to predict drug metabolism.
Results A comparative analysis was conducted between a CNN model utilizing solely the CYP2D6 genotype data and a multimodal model incorporating both genotype and drug-specific information. The multimodal approach demonstrated better performance across all evaluated metrics. An additional experiment predicting drug metabolism on unseen drug data also performed well.
Conclusions This model is anticipated to enhance the prediction of metabolic capacity in previously uncharacterized CYP2D6 variants, potentially reducing adverse drug reactions.
Methods A multimodal deep learning approach that combined CYP2D6 genotype data and drug structural information was used in this study. A Convolutional Neural Network (CNN) was used to encode the genotype data, and a Graph Convolutional Network (GCN) was used to decode the drug structures. These diverse data types were then integrated into a multimodal model to predict drug metabolism.
Results A comparative analysis was conducted between a CNN model utilizing solely the CYP2D6 genotype data and a multimodal model incorporating both genotype and drug-specific information. The multimodal approach demonstrated better performance across all evaluated metrics. An additional experiment predicting drug metabolism on unseen drug data also performed well.
Conclusions This model is anticipated to enhance the prediction of metabolic capacity in previously uncharacterized CYP2D6 variants, potentially reducing adverse drug reactions.
13.
μ΄λν; μ μ μ©
Abstract | Links | BibTeX | Tags: Cardiotoxicity, Machine learning
@conference{μ΄λν2024κΈ°κ³νμ΅,
title = {κΈ°κ³νμ΅ κΈ°λ° νν©λ¬Όμ μ¬μ₯λ
μ± μμΈ‘ μ°κ΅¬},
author = {μ΄λν and μ μ μ©},
url = {https://www.dbpia.co.kr/pdf/pdfView.do?nodeId=NODE11862000&googleIPSandBox=false&mark=0&minRead=5&ipRange=false&b2cLoginYN=false&icstClss=010000&isPDFSizeAllowed=true&accessgl=Y&language=ko_KR&hasTopBanner=true},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
booktitle = {νκ΅μ 보과νν νμ λ°νλ
Όλ¬Έμ§},
journal = {νκ΅μ 보과νν νμ λ°νλ
Όλ¬Έμ§},
pages = {825β827},
publisher = {νκ΅μ 보과νν},
abstract = {μΈκ° μν
λ₯΄-μ-κ³ -κ³ κ΄λ ¨ μ μ μ(hERG) μ±λμ μ¬μ₯μ μ κΈ°μ νλμ μ‘°μ νλ λ° μ€μν μν μ νλ€. μ΄ μ±λμ μ°¨λ¨νλ μ½λ¬Όμ μ¬κ°ν μ¬μ₯λ
μ±μ μΌμΌν¬ μ μλλ°, κΈ°μ‘΄μ μμ μ± κ²μ¬λ λ§μ μκ°κ³Ό λΉμ©μ μꡬνλ€λ λ¨μ μ΄ μλ€. μ΄ λ¬Έμ λ₯Ό ν΄κ²°νκΈ° μν΄, λ³Έ μ°κ΅¬μμλ in silico λ°©λ²μ μ΄μ©νμ¬ hERG μ°¨λ¨μ λ₯Ό μμΈ‘ν¨μΌλ‘μ¨ μ¬μ₯λ
μ±μ νμ
νλ λͺ¨λΈμ μ μνλ€. νν©λ¬Όμ ꡬ쑰μ μ 보λ₯Ό νμ
νκΈ° μν΄ ECFP(Extended Connectivity Fingerprint)λ₯Ό μ¬μ©νμ¬ λ³ννμκ³ . 물리ννμ νΉμ± λν μΆμΆνμκ³ , μΆμΆν λ°μ΄ν°λ₯Ό κΈ°λ°μΌλ‘ κΈ°κ³νμ΅ λͺ¨λΈμ ꡬμΆνμλ€. μ΄ μ κ·Όλ²μ μ¬μ₯λ
μ±μ μ λ°ν μ μλ μ μ½ ν보 λ¬Όμ§μ ν¨κ³Όμ μΌλ‘ μ λ³ν μ μκ² νλ€. κ²°κ³Όμ μΌλ‘, μ΄ μ°κ΅¬λ μμ νκ³ ν¨μ¨μ μΈ ν보 λ¬Όμ§μ λ°κ΅΄μ μ€μν κΈ°μ¬λ₯Ό ν κ²μΌλ‘ κΈ°λλλ€ },
keywords = {Cardiotoxicity, Machine learning},
pubstate = {published},
tppubtype = {conference}
}
μΈκ° μν
λ₯΄-μ-κ³ -κ³ κ΄λ ¨ μ μ μ(hERG) μ±λμ μ¬μ₯μ μ κΈ°μ νλμ μ‘°μ νλ λ° μ€μν μν μ νλ€. μ΄ μ±λμ μ°¨λ¨νλ μ½λ¬Όμ μ¬κ°ν μ¬μ₯λ
μ±μ μΌμΌν¬ μ μλλ°, κΈ°μ‘΄μ μμ μ± κ²μ¬λ λ§μ μκ°κ³Ό λΉμ©μ μꡬνλ€λ λ¨μ μ΄ μλ€. μ΄ λ¬Έμ λ₯Ό ν΄κ²°νκΈ° μν΄, λ³Έ μ°κ΅¬μμλ in silico λ°©λ²μ μ΄μ©νμ¬ hERG μ°¨λ¨μ λ₯Ό μμΈ‘ν¨μΌλ‘μ¨ μ¬μ₯λ
μ±μ νμ
νλ λͺ¨λΈμ μ μνλ€. νν©λ¬Όμ ꡬ쑰μ μ 보λ₯Ό νμ
νκΈ° μν΄ ECFP(Extended Connectivity Fingerprint)λ₯Ό μ¬μ©νμ¬ λ³ννμκ³ . 물리ννμ νΉμ± λν μΆμΆνμκ³ , μΆμΆν λ°μ΄ν°λ₯Ό κΈ°λ°μΌλ‘ κΈ°κ³νμ΅ λͺ¨λΈμ ꡬμΆνμλ€. μ΄ μ κ·Όλ²μ μ¬μ₯λ
μ±μ μ λ°ν μ μλ μ μ½ ν보 λ¬Όμ§μ ν¨κ³Όμ μΌλ‘ μ λ³ν μ μκ² νλ€. κ²°κ³Όμ μΌλ‘, μ΄ μ°κ΅¬λ μμ νκ³ ν¨μ¨μ μΈ ν보 λ¬Όμ§μ λ°κ΅΄μ μ€μν κΈ°μ¬λ₯Ό ν κ²μΌλ‘ κΈ°λλλ€
12.
λ°μ€μ; μ μ μ©
Abstract | Links | BibTeX | Tags: Network analysis
@conference{λ°μ€μ2024λ€νΈμν¬,
title = {λ€νΈμν¬ λΆμμ ν΅ν νν©λ¬Ό ννν ν¨κ³Ό μΆλ‘ },
author = {λ°μ€μ and μ μ μ©},
url = {https://www.dbpia.co.kr/pdf/pdfView.do?nodeId=NODE11861866&googleIPSandBox=false&mark=0&minRead=5&ipRange=false&b2cLoginYN=false&icstClss=010000&isPDFSizeAllowed=true&accessgl=Y&language=ko_KR&hasTopBanner=true},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
booktitle = {νκ΅μ 보과νν νμ λ°νλ
Όλ¬Έμ§},
journal = {νκ΅μ 보과νν νμ λ°νλ
Όλ¬Έμ§},
pages = {423β425},
publisher = {νκ΅μ 보과νν},
abstract = {μ½λ¬Όμ μμμΉ λͺ»ν λΆμμ©μ μ λ°ν μ μκΈ° λλ¬Έμ κ°λ°κ³Όμ μμ μ μ¬μ μΈ λΆμμ©μ μλ³νλ κ²μ΄ νμμ μ΄λ€. λ³Έ λ
Όλ¬Έμμλ μ½λ¬Όμ νμ± μ±λΆμΈ νν©λ¬Όμ μΈμ²΄μ λν μ μ¬μ μΈ λΆμμ©μ μλ³νλ λ° μμ΄μ λ€νΈμν¬ λΆμκ³Ό Random Walk with Restart(RWR) μκ³ λ¦¬μ¦μ λ³ννμ¬ νμ©νλ€. λ€μν νν©λ¬Όμ μν΄ μ λλ μ μλ νννμ μμΈ‘νκ³ , μ΄λ₯Ό ν΅ν΄ λ
μ±μ νκ°νλ μ κ·Όλ°©μμ μ§ννλ€. λ¨λ°±μ§ μνΈμμ© λ€νΈμν¬ κ΅¬μΆκ³Ό λΆμμ ν΅ν΄ νν©λ¬Όκ³Ό μ μ μ μνΈμμ©μ 볡μ‘μ±μ ν¬μ°©νκ³ μ μ¬μ μΈ λΆμμ©μ ν¨μ¨μ μΌλ‘ μλ³ν μ μλ€. λν νν©λ¬Όκ³Ό μ μ μ, νννκ°μ μ°κ΄μ± μ 보λ₯Ό νμ©νμ¬ νν©λ¬Όμ ν¨κ³Όλ₯Ό λμΆνκ³ , ν΅κ³μ κΈ°λ²μ νμ©νμ¬ μ λ’°μ± λμ νννμ μΆλ‘ ν μ μλ€. μ΄λ μ½λ¬Ό λ
μ± μμΈ‘κ³Ό μλ‘μ΄ μ½λ¬Ό νμ λ°κ²¬μ κΈ°μ¬ν μ μλ κ°λ₯μ±μ 보μ¬μ£Όλ©° μ½λ¬Ό μ€ν¬λ¦¬λ λ°©λ²μ κ°μ μ μ μλ―Έν μ 보λ₯Ό μ 곡νλ€},
keywords = {Network analysis},
pubstate = {published},
tppubtype = {conference}
}
μ½λ¬Όμ μμμΉ λͺ»ν λΆμμ©μ μ λ°ν μ μκΈ° λλ¬Έμ κ°λ°κ³Όμ μμ μ μ¬μ μΈ λΆμμ©μ μλ³νλ κ²μ΄ νμμ μ΄λ€. λ³Έ λ
Όλ¬Έμμλ μ½λ¬Όμ νμ± μ±λΆμΈ νν©λ¬Όμ μΈμ²΄μ λν μ μ¬μ μΈ λΆμμ©μ μλ³νλ λ° μμ΄μ λ€νΈμν¬ λΆμκ³Ό Random Walk with Restart(RWR) μκ³ λ¦¬μ¦μ λ³ννμ¬ νμ©νλ€. λ€μν νν©λ¬Όμ μν΄ μ λλ μ μλ νννμ μμΈ‘νκ³ , μ΄λ₯Ό ν΅ν΄ λ
μ±μ νκ°νλ μ κ·Όλ°©μμ μ§ννλ€. λ¨λ°±μ§ μνΈμμ© λ€νΈμν¬ κ΅¬μΆκ³Ό λΆμμ ν΅ν΄ νν©λ¬Όκ³Ό μ μ μ μνΈμμ©μ 볡μ‘μ±μ ν¬μ°©νκ³ μ μ¬μ μΈ λΆμμ©μ ν¨μ¨μ μΌλ‘ μλ³ν μ μλ€. λν νν©λ¬Όκ³Ό μ μ μ, νννκ°μ μ°κ΄μ± μ 보λ₯Ό νμ©νμ¬ νν©λ¬Όμ ν¨κ³Όλ₯Ό λμΆνκ³ , ν΅κ³μ κΈ°λ²μ νμ©νμ¬ μ λ’°μ± λμ νννμ μΆλ‘ ν μ μλ€. μ΄λ μ½λ¬Ό λ
μ± μμΈ‘κ³Ό μλ‘μ΄ μ½λ¬Ό νμ λ°κ²¬μ κΈ°μ¬ν μ μλ κ°λ₯μ±μ 보μ¬μ£Όλ©° μ½λ¬Ό μ€ν¬λ¦¬λ λ°©λ²μ κ°μ μ μ μλ―Έν μ 보λ₯Ό μ 곡νλ€
11.
μ‘μ€μ£Ό; μ μ μ©
Abstract | Links | BibTeX | Tags: Deep learning, Graph attention network
@conference{μ‘μ€μ£Ό2024νν©λ¬Όμ,
title = {νν©λ¬Όμ ν λ°μμ± μμΈ‘μ μν κ·Έλν μ κ²½λ§ μ κ·Όλ²},
author = {μ‘μ€μ£Ό and μ μ μ©},
url = {https://www.dbpia.co.kr/pdf/pdfView.do?nodeId=NODE11861976&googleIPSandBox=false&mark=0&minRead=5&ipRange=false&b2cLoginYN=false&icstClss=010000&isPDFSizeAllowed=true&accessgl=Y&language=ko_KR&hasTopBanner=true},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
booktitle = {νκ΅μ 보과νν νμ λ°νλ
Όλ¬Έμ§},
journal = {νκ΅μ 보과νν νμ λ°νλ
Όλ¬Έμ§},
pages = {753β755},
publisher = {νκ΅μ 보과νν},
abstract = {νμμ 맀λ
μλ°±λ§ λͺ
μ μ¬λ§μλ₯Ό μ΄λνλ μ£Όμ μ§ν μ€ νλμ΄λ©°, νΉν 2022λ
νκ΅μμλ μ μ€ μ¬λ§λ₯ μ΄ κ°μ₯ λμ μ§νμΌλ‘ κΈ°λ‘λμλ€. μ΄μ λ°λΌ, νμμ μ λ°νλ νν©λ¬Όμ λν μ΄ν΄μ μ°κ΅¬κ° νμμ μ΄λ©°, λ³Έ μ°κ΅¬λ κΈ°μ‘΄μ κΈ°κ³νμ΅ λ° λ₯λ¬λ λ°©λ²μ νκ³λ₯Ό 극볡νκ³ , νν©λ¬Όμ νμ μ λ° κ°λ₯μ±μ μμΈ‘νκΈ° μν΄ Graph Attention Network (GAT)λ₯Ό νμ©ν μλ‘μ΄ μ κ·Όλ°©μμ μ μνκ³ νκ°νμλ€. λ³Έ μ°κ΅¬μμλ νν©λ¬Ό λ°μμ± λ°μ΄ν°μΈ CPDBμ CCRIS λ°μ΄ν°λ² μ΄μ€λ₯Ό νμ©νμμΌλ©°, Simplified Molecular Input Line Entry System (SMILES) μ 보λ₯Ό κΈ°λ°μΌλ‘ λΆμμ ꡬ쑰μ ννμ μ±μ§μ κ·Έλν λ°μ΄ν°λ‘ λ³ννμλ€. GAT λͺ¨λΈμ μ΄ κ·Έλν λ°μ΄ν°λ₯Ό μ΄μ©νμ¬ λΆμ κ°μ 볡μ‘ν μνΈμμ©μ νμ΅νκ³ , νμ λ°μ κ°λ₯μ±μ μμΈ‘νμμΌλ©°, μ±λ₯ νκ°μμ λ€λ₯Έ λͺ¨λΈκ³Ό λΉκ΅νμ¬ κ°μ₯ μ°μν μμΈ‘ μ±λ₯μ μ
μ¦νμλ€. μ΄λ νμ μμΈ‘μ μν ν¨κ³Όμ μΈ λꡬλ‘μ GATμ μ μ¬λ ₯μ 보μ¬μ£Όλ©°, ν₯ν μ μ°κ΅¬ λ° μΉλ£ κ°λ°μ μ€μν κΈ°μ¬λ₯Ό ν },
keywords = {Deep learning, Graph attention network},
pubstate = {published},
tppubtype = {conference}
}
νμμ 맀λ
μλ°±λ§ λͺ
μ μ¬λ§μλ₯Ό μ΄λνλ μ£Όμ μ§ν μ€ νλμ΄λ©°, νΉν 2022λ
νκ΅μμλ μ μ€ μ¬λ§λ₯ μ΄ κ°μ₯ λμ μ§νμΌλ‘ κΈ°λ‘λμλ€. μ΄μ λ°λΌ, νμμ μ λ°νλ νν©λ¬Όμ λν μ΄ν΄μ μ°κ΅¬κ° νμμ μ΄λ©°, λ³Έ μ°κ΅¬λ κΈ°μ‘΄μ κΈ°κ³νμ΅ λ° λ₯λ¬λ λ°©λ²μ νκ³λ₯Ό 극볡νκ³ , νν©λ¬Όμ νμ μ λ° κ°λ₯μ±μ μμΈ‘νκΈ° μν΄ Graph Attention Network (GAT)λ₯Ό νμ©ν μλ‘μ΄ μ κ·Όλ°©μμ μ μνκ³ νκ°νμλ€. λ³Έ μ°κ΅¬μμλ νν©λ¬Ό λ°μμ± λ°μ΄ν°μΈ CPDBμ CCRIS λ°μ΄ν°λ² μ΄μ€λ₯Ό νμ©νμμΌλ©°, Simplified Molecular Input Line Entry System (SMILES) μ 보λ₯Ό κΈ°λ°μΌλ‘ λΆμμ ꡬ쑰μ ννμ μ±μ§μ κ·Έλν λ°μ΄ν°λ‘ λ³ννμλ€. GAT λͺ¨λΈμ μ΄ κ·Έλν λ°μ΄ν°λ₯Ό μ΄μ©νμ¬ λΆμ κ°μ 볡μ‘ν μνΈμμ©μ νμ΅νκ³ , νμ λ°μ κ°λ₯μ±μ μμΈ‘νμμΌλ©°, μ±λ₯ νκ°μμ λ€λ₯Έ λͺ¨λΈκ³Ό λΉκ΅νμ¬ κ°μ₯ μ°μν μμΈ‘ μ±λ₯μ μ
μ¦νμλ€. μ΄λ νμ μμΈ‘μ μν ν¨κ³Όμ μΈ λꡬλ‘μ GATμ μ μ¬λ ₯μ 보μ¬μ£Όλ©°, ν₯ν μ μ°κ΅¬ λ° μΉλ£ κ°λ°μ μ€μν κΈ°μ¬λ₯Ό ν
10.
μλ¬ΈμλΉ; μ μ μ©
Abstract | Links | BibTeX | Tags: CYP450, Deep learning, Graph attention network
@conference{μλ¬ΈμλΉ2024cytochrome,
title = {Cytochrome P450 λμ체 μ΅μ μ μμΈ‘μ μν κ·Έλν μ΄ν
μ
λ€νΈμν¬ λͺ¨λΈ κ°λ°},
author = {μλ¬ΈμλΉ and μ μ μ©},
url = {https://www.dbpia.co.kr/pdf/pdfView.do?nodeId=NODE11861993&googleIPSandBox=false&mark=0&minRead=5&ipRange=false&b2cLoginYN=false&icstClss=010000&isPDFSizeAllowed=true&accessgl=Y&language=ko_KR&hasTopBanner=true},
year = {2024},
date = {2024-01-01},
urldate = {2024-01-01},
booktitle = {νκ΅μ 보과νν νμ λ°νλ
Όλ¬Έμ§},
journal = {νκ΅μ 보과νν νμ λ°νλ
Όλ¬Έμ§},
pages = {804β806},
publisher = {νκ΅μ 보과νν},
abstract = {Cytochrome P450 ν¨μλ λͺ¨λ λμ¬ λ°μ μ€ μ½ 75%λ₯Ό μ±
μμ§λ©°, νΉν 1A2, 2C9, 2C19, 2D6, 3A4 λ±μ λλ€μ μ½λ¬Όμ λμ¬μ κ΄μ¬νκ³ , λ€μμ λΆμμ©μ μ λ°νλ κ²μΌλ‘ μλ €μ Έ μλ€. μ΄μ λ°λΌ, μ μ½ κ°λ° κ³Όμ μμ μ΄λ€ cytochrome P450μ μ΅μ νλ νν©λ¬Όμ μλ³νλ κ²μ λ§€μ° μ€μνλ€. λ³Έ λ
Όλ¬Έμ μ½λ¬Ό λΆμμ κ·Έλν ꡬ쑰λ₯Ό μ΄μ©νκ³ self-attention λ©μ»€λμ¦μ μ μ©νμ¬ P450 λμ체λ₯Ό μ΅μ νλ νν©λ¬Όμ μμΈ‘νλ μλ‘μ΄ λͺ¨λΈμ μ μνλ€. μ΄ λͺ¨λΈμ Graph Attention Network (GAT)λ₯Ό νμ©νμ¬ λΆμμ κ·Έλν ννμ νμ΅νκ³ , Fully-connected layerμ ν΅ν΄ μμΈ‘μ μννλ€. λν, λ°μ΄ν°μ λΆκ· ν λ¬Έμ λ₯Ό ν΄κ²°νκΈ° μν΄ Focal loss ν¨μλ₯Ό μ μ©νμλ€. μ΄ μ°κ΅¬λ in vivoμ λλ λΉμ©κ³Ό μκ°μ μ κ°νκ³ , μ μ½ κ°λ°μ κΈ°κ°κ³Ό λΉμ©μ μ€μ΄λλ° κΈ°μ¬ν κ²μΌλ‘ κΈ°λλλ€},
keywords = {CYP450, Deep learning, Graph attention network},
pubstate = {published},
tppubtype = {conference}
}
Cytochrome P450 ν¨μλ λͺ¨λ λμ¬ λ°μ μ€ μ½ 75%λ₯Ό μ±
μμ§λ©°, νΉν 1A2, 2C9, 2C19, 2D6, 3A4 λ±μ λλ€μ μ½λ¬Όμ λμ¬μ κ΄μ¬νκ³ , λ€μμ λΆμμ©μ μ λ°νλ κ²μΌλ‘ μλ €μ Έ μλ€. μ΄μ λ°λΌ, μ μ½ κ°λ° κ³Όμ μμ μ΄λ€ cytochrome P450μ μ΅μ νλ νν©λ¬Όμ μλ³νλ κ²μ λ§€μ° μ€μνλ€. λ³Έ λ
Όλ¬Έμ μ½λ¬Ό λΆμμ κ·Έλν ꡬ쑰λ₯Ό μ΄μ©νκ³ self-attention λ©μ»€λμ¦μ μ μ©νμ¬ P450 λμ체λ₯Ό μ΅μ νλ νν©λ¬Όμ μμΈ‘νλ μλ‘μ΄ λͺ¨λΈμ μ μνλ€. μ΄ λͺ¨λΈμ Graph Attention Network (GAT)λ₯Ό νμ©νμ¬ λΆμμ κ·Έλν ννμ νμ΅νκ³ , Fully-connected layerμ ν΅ν΄ μμΈ‘μ μννλ€. λν, λ°μ΄ν°μ λΆκ· ν λ¬Έμ λ₯Ό ν΄κ²°νκΈ° μν΄ Focal loss ν¨μλ₯Ό μ μ©νμλ€. μ΄ μ°κ΅¬λ in vivoμ λλ λΉμ©κ³Ό μκ°μ μ κ°νκ³ , μ μ½ κ°λ°μ κΈ°κ°κ³Ό λΉμ©μ μ€μ΄λλ° κΈ°μ¬ν κ²μΌλ‘ κΈ°λλλ€
2023
9.
Myeonghyeon Jeong; Sunyong Yoo
Links | BibTeX | Tags: Machine learning
@conference{nokey,
title = {FetoML: Interpretable predictions of the fetotoxicity of drugs based on machine learning approaches},
author = {Myeonghyeon Jeong and Sunyong Yoo},
url = {https://dtmbio.net/},
year = {2023},
date = {2023-01-02},
urldate = {2023-01-02},
booktitle = {In 17th International Conference on Data and Text Mining in Biomedical Informatics},
pages = {20},
publisher = {DTMBIO},
keywords = {Machine learning},
pubstate = {published},
tppubtype = {conference}
}
8.
Dohyeon Lee; Sunyong Yoo
Links | BibTeX | Tags: Deep learning, Graph attention network
@conference{nokey,
title = {hERGAT: Predicting hERG blockers using graph attention mechanism through atom- and molecule- level interaction analysis},
author = {Dohyeon Lee and Sunyong Yoo},
url = {https://dtmbio.net/},
year = {2023},
date = {2023-01-02},
urldate = {2023-01-02},
booktitle = {In 17th International Conference on Data and Text Mining in Biomedical Informatics},
publisher = {DTMBIO},
keywords = {Deep learning, Graph attention network},
pubstate = {published},
tppubtype = {conference}
}
2022
7.
Myeonghyeon Jeong; Sangjin Kim; Yewon Han; Jihyun Jeong; Dahwa Jung; Inyoung Choi; Sunyong Yoo
BibTeX | Tags: Attention mechanism, Bioinformatics, Deep learning
@conference{nokey,
title = {Attention-based Deep Neural Network for Predicting Fetotoxicity},
author = {Myeonghyeon Jeong and Sangjin Kim and Yewon Han and Jihyun Jeong and Dahwa Jung and Inyoung Choi and Sunyong Yoo},
year = {2022},
date = {2022-01-02},
urldate = {2022-01-02},
booktitle = {In the 10th International Conference on Big Data Applications and Services},
publisher = {The Korea Big Data Service Society},
keywords = {Attention mechanism, Bioinformatics, Deep learning},
pubstate = {published},
tppubtype = {conference}
}
6.
μ μ μ°; μ μ μ©
Links | BibTeX | Tags: DDI, Text mining
@conference{μ μ μ°2022μ½λ¬Ό,
title = {μ½λ¬Ό μ 보 λ¬Έμ μλ² λ©μ μ΄μ©ν λ₯λ¬λ κΈ°λ° μ½λ¬Ό κ° μνΈμμ© μμΈ‘},
author = {μ μ μ° and μ μ μ©},
url = {https://koreascience.kr/article/CFKO202221536102022.pdf},
year = {2022},
date = {2022-01-01},
urldate = {2022-01-01},
booktitle = {νκ΅μ 보ν΅μ νν μ’
ν©νμ λν λ
Όλ¬Έμ§},
journal = {νκ΅μ 보ν΅μ νν μ’
ν©νμ λν λ
Όλ¬Έμ§},
volume = {26},
number = {1},
pages = {276β278},
publisher = {νκ΅μ 보ν΅μ νν},
keywords = {DDI, Text mining},
pubstate = {published},
tppubtype = {conference}
}
5.
μ΄μμ°; μ μ μ©
Links | BibTeX | Tags: in silico
@conference{μ΄μμ°2022silico,
title = {In silico κΈ°λ²μ μ΄μ©ν μ κ²½λ
μ± μμΈ‘},
author = {μ΄μμ° and μ μ μ©},
url = {https://www.dbpia.co.kr/pdf/pdfView.do?nodeId=NODE11077893&googleIPSandBox=false&mark=0&minRead=5&ipRange=false&b2cLoginYN=false&icstClss=010000&isPDFSizeAllowed=true&accessgl=Y&language=ko_KR&hasTopBanner=true},
year = {2022},
date = {2022-01-01},
urldate = {2022-01-01},
booktitle = {νκ΅μ 보ν΅μ νν μ’
ν©νμ λν λ
Όλ¬Έμ§},
journal = {νκ΅μ 보ν΅μ νν μ’
ν©νμ λν λ
Όλ¬Έμ§},
volume = {26},
number = {1},
pages = {270β272},
publisher = {νκ΅μ 보ν΅μ νν},
keywords = {in silico},
pubstate = {published},
tppubtype = {conference}
}
4.
μ λͺ
ν; μ μ μ©
Links | BibTeX | Tags: Attention mechanism
@conference{μ λͺ
ν2022attention,
title = {Attention μκ³ λ¦¬μ¦ κΈ°λ° μ½λ¬Όμ νμ λ
μ± μμΈ‘ μ°κ΅¬},
author = {μ λͺ
ν and μ μ μ©},
url = {https://www.dbpia.co.kr/pdf/pdfView.do?nodeId=NODE11077894&googleIPSandBox=false&mark=0&minRead=5&ipRange=false&b2cLoginYN=false&icstClss=010000&isPDFSizeAllowed=true&accessgl=Y&language=ko_KR&hasTopBanner=true},
year = {2022},
date = {2022-01-01},
urldate = {2022-01-01},
booktitle = {νκ΅μ 보ν΅μ νν μ’
ν©νμ λν λ
Όλ¬Έμ§},
journal = {νκ΅μ 보ν΅μ νν μ’
ν©νμ λν λ
Όλ¬Έμ§},
volume = {26},
number = {1},
pages = {273β275},
publisher = {νκ΅μ 보ν΅μ νν},
keywords = {Attention mechanism},
pubstate = {published},
tppubtype = {conference}
}
2015
3.
Moonshik Shin; Sungyoung Yoo; Suhyun Ha; Kyungrin Noh; Doheon Lee
Abstract | Links | BibTeX | Dimensions | Tags: ADME, Bioinformatics, Natural product
@conference{shin2015identifying,
title = {Identifying Potential Bioactive Compounds of Natural Products by Combining ADMET Prediction Methods},
author = {Moonshik Shin and Sungyoung Yoo and Suhyun Ha and Kyungrin Noh and Doheon Lee},
url = {https://dl.acm.org/doi/abs/10.1145/2811163.2811168},
doi = {10.1145/2811163.2811168},
year = {2015},
date = {2015-01-01},
urldate = {2015-01-01},
booktitle = {Proceedings of the ACM Ninth International Workshop on Data and Text Mining in Biomedical Informatics},
pages = {19β19},
publisher = {CIKM},
abstract = {Herbs consist of various chemical compounds. Thus, identifying potential bioactive compounds from those diversity is an important task for studies in the herb, food and natural products. Even though various computational approaches are developed for predicting bioactive compounds, the prediction performances are diverse due to different methods and dataset. Therefore, there is urgent demand for an approach that connotes the previous methods and identify potential bioactive compounds with high accuracy. To meet the demand, we proposed a filtering strategy that identifies potential bioactive compounds by combining previously developed computational methods which predict ADMET, such as Human Intestinal Absorption (HIA) and Caco-2 permeability. Our approach was evaluated on 930 compounds that are known as bioactive compounds, which were extracted from literature, DrugBank and Dr. Dukes phytochemical databases. By applying our filtering strategy, 97.5% of the known bioactive compounds were correctly predicted as bioactive. We examined whether our approach can distinguish the potential bioactive compound from the non-potential bioactive compounds with Fishers' exact test, and a reasonable p-value (3.806 x 10-9) was derived. For the next step, we are planning to develop a machine-learning based method to improve our filtering approach.},
keywords = {ADME, Bioinformatics, Natural product},
pubstate = {published},
tppubtype = {conference}
}
Herbs consist of various chemical compounds. Thus, identifying potential bioactive compounds from those diversity is an important task for studies in the herb, food and natural products. Even though various computational approaches are developed for predicting bioactive compounds, the prediction performances are diverse due to different methods and dataset. Therefore, there is urgent demand for an approach that connotes the previous methods and identify potential bioactive compounds with high accuracy. To meet the demand, we proposed a filtering strategy that identifies potential bioactive compounds by combining previously developed computational methods which predict ADMET, such as Human Intestinal Absorption (HIA) and Caco-2 permeability. Our approach was evaluated on 930 compounds that are known as bioactive compounds, which were extracted from literature, DrugBank and Dr. Dukes phytochemical databases. By applying our filtering strategy, 97.5% of the known bioactive compounds were correctly predicted as bioactive. We examined whether our approach can distinguish the potential bioactive compound from the non-potential bioactive compounds with Fishers' exact test, and a reasonable p-value (3.806 x 10-9) was derived. For the next step, we are planning to develop a machine-learning based method to improve our filtering approach.
2014
2.
Suhyun Ha; Sunyong Yoo; Moonshik Shin; Jin Sook Kwak; Oran Kwon; Min Chang Choi; Keon Wook Kang; Hojung Nam; Doheon Lee
Abstract | Links | BibTeX | Dimensions | Tags: Ethnopharmacology, Natural product
@conference{ha2014integrative,
title = {Integrative Database for Exploring Compound Combinations of Natural Products for Medical Effects},
author = {Suhyun Ha and Sunyong Yoo and Moonshik Shin and Jin Sook Kwak and Oran Kwon and Min Chang Choi and Keon Wook Kang and Hojung Nam and Doheon Lee},
url = {https://dl.acm.org/doi/abs/10.1145/2665970.2665986},
doi = {10.1145/2665970.2665986},
year = {2014},
date = {2014-01-01},
urldate = {2014-01-01},
booktitle = {Proceedings of the ACM 8th International Workshop on Data and Text Mining in Bioinformatics},
pages = {41β41},
publisher = {CIKM},
abstract = {Natural products used in dietary supplements, complementary and alternative medicine (CAM) and conventional medicine are composites of multiple chemical compounds. These chemical compounds potentially offer an extensive source for drug discovery with accumulated knowledge of efficacy and safety. However, existing natural product related databases have drawbacks in both standardization and structuralization of information. Therefore, in this work, we construct an integrated database of natural products by mapping the prescription, herb, compound, and phenotype information to international identifiers and structuralizing the efficacy information through database integration and text-mining methods. We expect that the constructed database could serve as a fundamental resource for the natural products research.},
keywords = {Ethnopharmacology, Natural product},
pubstate = {published},
tppubtype = {conference}
}
Natural products used in dietary supplements, complementary and alternative medicine (CAM) and conventional medicine are composites of multiple chemical compounds. These chemical compounds potentially offer an extensive source for drug discovery with accumulated knowledge of efficacy and safety. However, existing natural product related databases have drawbacks in both standardization and structuralization of information. Therefore, in this work, we construct an integrated database of natural products by mapping the prescription, herb, compound, and phenotype information to international identifiers and structuralizing the efficacy information through database integration and text-mining methods. We expect that the constructed database could serve as a fundamental resource for the natural products research.
2012
1.
Moonshik Shin; Sunyong Yoo; Kwang H Lee; Doheon Lee
Abstract | Links | BibTeX | Dimensions | Tags:
@conference{shin2012electronic,
title = {Electronic medical records privacy preservation through k-anonymity clustering method},
author = {Moonshik Shin and Sunyong Yoo and Kwang H Lee and Doheon Lee},
url = {https://ieeexplore.ieee.org/abstract/document/6505046},
doi = {10.1109/SCIS-ISIS.2012.6505046},
year = {2012},
date = {2012-01-01},
urldate = {2012-01-01},
booktitle = {The 6th International Conference on Soft Computing and Intelligent Systems, and The 13th International Symposium on Advanced Intelligence Systems},
pages = {1119β1124},
publisher = {IEEE},
organization = {IEEE},
abstract = {Electronic Medical Records (EMRs) enable the sharing of patient medical data whenever it is needed and also are used as a tool for building new medical technology and patient recommendation systems. Since EMRs include patients' private data, access is restricted to researchers. Thus, an anonymizing technique is necessary that keeps patients' private data safe while not damaging useful medical information. k-member clustering anonymization approaches k-anonymization as a clustering issue. The objective of the k-member clustering problem is to gather records that will minimize the data distortion during data generalization. Most of the previous clustering techniques include random seed selection. However, randomly selecting a cluster seed will provide inconsistent performance. The authors propose a k-member cluster seed selection algorithm (KMCSSA) that is distinct from the previous clustering methods. Instead of randomly selecting a cluster seed, the proposed method selects the seed based on the closeness centrality to provide consistent information loss (IL) and to reduce the information distortion. An adult database from University of California Irvine Machine Learning Repository was used for the experiment. By comparing the proposed method with two previous methods, the experimental results shows that KMCSSA provides superior performance with respect to information loss. The authors provide a privacy protection algorithm that derives consistent information loss and reduces the overall information distortion.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Electronic Medical Records (EMRs) enable the sharing of patient medical data whenever it is needed and also are used as a tool for building new medical technology and patient recommendation systems. Since EMRs include patients' private data, access is restricted to researchers. Thus, an anonymizing technique is necessary that keeps patients' private data safe while not damaging useful medical information. k-member clustering anonymization approaches k-anonymization as a clustering issue. The objective of the k-member clustering problem is to gather records that will minimize the data distortion during data generalization. Most of the previous clustering techniques include random seed selection. However, randomly selecting a cluster seed will provide inconsistent performance. The authors propose a k-member cluster seed selection algorithm (KMCSSA) that is distinct from the previous clustering methods. Instead of randomly selecting a cluster seed, the proposed method selects the seed based on the closeness centrality to provide consistent information loss (IL) and to reduce the information distortion. An adult database from University of California Irvine Machine Learning Repository was used for the experiment. By comparing the proposed method with two previous methods, the experimental results shows that KMCSSA provides superior performance with respect to information loss. The authors provide a privacy protection algorithm that derives consistent information loss and reduces the overall information distortion.