Junghyun Namkung
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Junghyun Namkung has a total of 50 co-authors. Here are the top co-authors by number of shared publications:
- Taesung Park
Interdisciplinary Program in Bioinformatics, Seoul National University, Seoul, Republic of Korea
5 shared publications - Jin-Young Jang
Department of Surgery and Cancer Research Institute, Seoul National University College of Medicine, Seoul, Republic of Korea
2 shared publications - Seungyeoun Lee
Department of Mathematics and Statistics, Sejong University, Seoul 05006, Korea
2 shared publications - Wooil Kwon
Department of Surgery and Cancer Research Institute, Seoul National University College of Medicine , Seoul
2 shared publications - Yongkang Kim
Department of Statistics, Seoul National University, Seoul, Republic of Korea
2 shared publications - Areum Lee
Biomedical Research Institute, Seoul National University Hospital, Seoul, Korea
1 shared publication
Latest Publications
- Open Access
- 1 January 2026
- Published by XMLink in Annals of Dermatology
- Vol. 38
Comprehensive Analysis of Factors Affecting Facial Skin Biophysical Profiles in Korean: A Cross-sectional Study
Oh S, et al. Ann Dermatol. 2026 ;38:e76. https://doi.org/10.5021/ad.26.119
- 1 January 2026
- Published by XMLink in Annals of Dermatology
- Vol. 38
Open AccessComprehensive Analysis of Factors Affecting Facial Skin Biophysical Profiles in Korean: A Cross-sectional Study
Oh S, et al. Ann Dermatol. 2026 ;38:e76. https://doi.org/10.5021/ad.26.119
- Open Access
- 25 November 2024
- Published by Korean Neuropsychiatric Association in Psychiatry Investigation
- Vol. 21 (11) , 1228-1237
Novel Deep Learning-Based Vocal Biomarkers for Stress Detection in Koreans
Objective The rapid societal changes have underscored the importance of effective stress detection and management. Chronic mental stress significantly contributes to both physical and psychological illnesses. However, many individuals often remain unaware of their stress levels until they face physical health issues, highlighting the necessity for regular stress monitoring. This study aimed to investigate the effectiveness of vocal biomarkers in detecting stress levels among healthy Korean employees and to contribute to digital healthcare solutions.Methods We conducted a multi-center clinical study by collecting voice recordings from 115 healthy Korean employees under both relaxed and stress-induced conditions. Stress was induced using the socially evaluated cold pressor test. The Emphasized Channel Attention, Propagation and Aggregation in Time delay neural network (ECAPA-TDNN) deep learning architecture, renowned for its advanced capabilities in analyzing person-specific voice features, was employed to develop stress prediction scores.Results The proposed model achieved a 70% accuracy rate in detecting stress. This performance underscores the potential of vocal biomarkers as a convenient and effective tool for individuals to self-monitor and manage their stress levels within digital healthcare frameworks.Conclusion The findings emphasize the promise of voice-based mental stress assessments within the Korean population and the importance of continued research on vocal biomarkers across diverse linguistic demographics.
- 25 November 2024
- Published by Korean Neuropsychiatric Association in Psychiatry Investigation
- Vol. 21 (11) , 1228-1237
Open AccessNovel Deep Learning-Based Vocal Biomarkers for Stress Detection in Koreans
Objective The rapid societal changes have underscored the importance of effective stress detection and management. Chronic mental stress significantly contributes to both physical and psychological illnesses. However, many individuals often remain unaware of their stress levels until they face physical health issues, highlighting the necessity for regular stress monitoring. This study aimed to investigate the effectiveness of vocal biomarkers in detecting stress levels among healthy Korean employees and to contribute to digital healthcare solutions.Methods We conducted a multi-center clinical study by collecting voice recordings from 115 healthy Korean employees under both relaxed and stress-induced conditions. Stress was induced using the socially evaluated cold pressor test. The Emphasized Channel Attention, Propagation and Aggregation in Time delay neural network (ECAPA-TDNN) deep learning architecture, renowned for its advanced capabilities in analyzing person-specific voice features, was employed to develop stress prediction scores.Results The proposed model achieved a 70% accuracy rate in detecting stress. This performance underscores the potential of vocal biomarkers as a convenient and effective tool for individuals to self-monitor and manage their stress levels within digital healthcare frameworks.Conclusion The findings emphasize the promise of voice-based mental stress assessments within the Korean population and the importance of continued research on vocal biomarkers across diverse linguistic demographics.
- Open Access
- 1 January 2021
- Published by The Korean Surgical Society in Annals of Surgical Treatment and Research
- Vol. 100 (3) , 144-153
Diagnostic model for pancreatic cancer using a multi-biomarker panel
Choi YJ, et al. Ann Surg Treat Res. 2021 Mar;100(3):144-153. https://doi.org/10.4174/astr.2021.100.3.144
- 1 January 2021
- Published by The Korean Surgical Society in Annals of Surgical Treatment and Research
- Vol. 100 (3) , 144-153
Open AccessDiagnostic model for pancreatic cancer using a multi-biomarker panel
Choi YJ, et al. Ann Surg Treat Res. 2021 Mar;100(3):144-153. https://doi.org/10.4174/astr.2021.100.3.144
- 27 February 2020
- Published by The Microbiological Society of Korea in Journal of Microbiology
- Vol. 58 (3) , 206-216
Machine learning methods for microbiome studies
Researches on the microbiome have been actively conducted worldwide and the results have shown human gut bacterial environment significantly impacts on immune system, psychological conditions, cancers, obesity, and metabolic diseases. Thanks to the development of sequencing technology, microbiome studies with large number of samples are eligible on an acceptable cost nowadays. Large samples allow analysis of more sophisticated modeling using machine learning approaches to study relationships between microbiome and various traits. This article provides an overview of machine learning methods for non-data scientists interested in the association analysis of microbiomes and host phenotypes. Once genomic feature of microbiome is determined, various analysis methods can be used to explore the relationship between microbiome and host phenotypes that include penalized regression, support vector machine (SVM), random forest, and artificial neural network (ANN). Deep neural network methods are also touched. Analysis procedure from environment setup to extract analysis results are presented with Python programming language.
- 27 February 2020
- Published by The Microbiological Society of Korea in Journal of Microbiology
- Vol. 58 (3) , 206-216
Machine learning methods for microbiome studies
Researches on the microbiome have been actively conducted worldwide and the results have shown human gut bacterial environment significantly impacts on immune system, psychological conditions, cancers, obesity, and metabolic diseases. Thanks to the development of sequencing technology, microbiome studies with large number of samples are eligible on an acceptable cost nowadays. Large samples allow analysis of more sophisticated modeling using machine learning approaches to study relationships between microbiome and various traits. This article provides an overview of machine learning methods for non-data scientists interested in the association analysis of microbiomes and host phenotypes. Once genomic feature of microbiome is determined, various analysis methods can be used to explore the relationship between microbiome and host phenotypes that include penalized regression, support vector machine (SVM), random forest, and artificial neural network (ANN). Deep neural network methods are also touched. Analysis procedure from environment setup to extract analysis results are presented with Python programming language.
- 31 October 2018
- Published by Springer Nature
- Vol. 1882, 261-286
Statistical Methods for Identifying Biomarkers from miRNA Profiles of Cancers
Biomarkers play important roles in early diagnosis and treatment plan for cancer patients and the importance is growing. With advances in high-throughput molecular profiling technology for various types of molecules such as DNA, RNA, proteins, or metabolites, it is now possible to perform massive profiling analysis that allows accelerating discovery of novel biomolecules. Because no single marker is sufficiently accurate for clinical use, the cancer biomarker is developed in the form of multiple biomarker panels. No single marker is sufficiently accurate for clinical use, and thus cancer biomarkers are developed in the form of multiple biomarker panels. Of various types of molecular biomarkers, microRNA (miRNA) has emerged as a class of promising cancer biomarker recently. MiRNAs are small noncoding RNAs that regulate gene expression. The chapter overviews the process of identification of biomarker panels from miRNA profiles focusing on statistical methods. Introduction to molecular cancer biomarkers is touched first. From sample design to miRNA profiling process is reviewed in the method section. Statistical methods for biomarker development are introduced according to three typical purposes of molecular biomarkers: tumor subtype classification, early detection, and prediction of treatment response or prognosis of patients. Example codes for R program are provided as well for selected methods.
- 31 October 2018
- Published by Springer Nature
- Vol. 1882, 261-286
Statistical Methods for Identifying Biomarkers from miRNA Profiles of Cancers
Biomarkers play important roles in early diagnosis and treatment plan for cancer patients and the importance is growing. With advances in high-throughput molecular profiling technology for various types of molecules such as DNA, RNA, proteins, or metabolites, it is now possible to perform massive profiling analysis that allows accelerating discovery of novel biomolecules. Because no single marker is sufficiently accurate for clinical use, the cancer biomarker is developed in the form of multiple biomarker panels. No single marker is sufficiently accurate for clinical use, and thus cancer biomarkers are developed in the form of multiple biomarker panels. Of various types of molecular biomarkers, microRNA (miRNA) has emerged as a class of promising cancer biomarker recently. MiRNAs are small noncoding RNAs that regulate gene expression. The chapter overviews the process of identification of biomarker panels from miRNA profiles focusing on statistical methods. Introduction to molecular cancer biomarkers is touched first. From sample design to miRNA profiling process is reviewed in the method section. Statistical methods for biomarker development are introduced according to three typical purposes of molecular biomarkers: tumor subtype classification, early detection, and prediction of treatment response or prognosis of patients. Example codes for R program are provided as well for selected methods.
- 5 October 2017
- Published by Springer Nature
- Vol. 1666, 409-439
Single Marker Family-Based Association Analysis Not Conditional on Parental Information
Family-based association analysis unconditional on parental genotypes models the effects of observed genotypes. This approach has been shown to have greater power than conditional methods. In this chapter, we review popular association analysis methods accounting for familial correlations: the marginal model using generalized estimating equations (GEE), the mixed model with a polygenic random component, and genome-wide association analyses. The marginal approach does not explicitly model familial correlations but uses the information to improve the efficiency of parameter estimates. This model, using GEE, is useful when the correlation structure is not of interest; the correlations are treated as nuisance parameters. In the mixed model, familial correlations are modeled as random effects, e.g., the polygenic inheritance model accounts for correlations originating from shared genomic components within a family. These unconditional methods provide a flexible modeling framework for general pedigree data to accommodate traits with various distributions and many types of covariate effects. Genome-wide association studies usually test more than 10,000 SNPs and thus traditional statistical methods accounting for the familial correlations often suffer from a computational burden. Multiple approaches that have been recently proposed to avoid this computational issue are reviewed. The single-marker analysis procedures are demonstrated using the R package gee and the ASSOC program in the S.A.G.E. package, including how to prepare input data, conduct the analysis, and interpret the output. ASSOC allows models to include random components of additional familial correlations that may be not sufficiently explained by a polygenic effect and addresses nonnormality of response variables by transformation methods. With its ease of use, ASSOC provides a useful tool for association analysis of large pedigree data.
- 5 October 2017
- Published by Springer Nature
- Vol. 1666, 409-439
Single Marker Family-Based Association Analysis Not Conditional on Parental Information
Family-based association analysis unconditional on parental genotypes models the effects of observed genotypes. This approach has been shown to have greater power than conditional methods. In this chapter, we review popular association analysis methods accounting for familial correlations: the marginal model using generalized estimating equations (GEE), the mixed model with a polygenic random component, and genome-wide association analyses. The marginal approach does not explicitly model familial correlations but uses the information to improve the efficiency of parameter estimates. This model, using GEE, is useful when the correlation structure is not of interest; the correlations are treated as nuisance parameters. In the mixed model, familial correlations are modeled as random effects, e.g., the polygenic inheritance model accounts for correlations originating from shared genomic components within a family. These unconditional methods provide a flexible modeling framework for general pedigree data to accommodate traits with various distributions and many types of covariate effects. Genome-wide association studies usually test more than 10,000 SNPs and thus traditional statistical methods accounting for the familial correlations often suffer from a computational burden. Multiple approaches that have been recently proposed to avoid this computational issue are reviewed. The single-marker analysis procedures are demonstrated using the R package gee and the ASSOC program in the S.A.G.E. package, including how to prepare input data, conduct the analysis, and interpret the output. ASSOC allows models to include random components of additional familial correlations that may be not sufficiently explained by a polygenic effect and addresses nonnormality of response variables by transformation methods. With its ease of use, ASSOC provides a useful tool for association analysis of large pedigree data.
- 1 January 2016
- Published by Inderscience Publishers in International Journal of Data Mining and Bioinformatics
- Vol. 16 (1) , 64
Comparative studies for developing protein based cancer prediction model to maximise the ROC-AUC with various variable selection methods
The era of protein data analysis is coming with more accurate quantification experiments such as the multiple reaction monitoring (MRM). Protein is easier to obtain than the other genetic variants or gene expression data, which makes it more suitable for early diagnosis of cancer. Each patient has unique patterns of protein data, which makes it imperative for the researcher to select the effective markers to construct a consistent model to predict the patients. This research focuses on finding the most effective variable selection method to be applied in the early diagnosis of the pancreatic cancer. In the process, we compare classical selection methods (stepwise selection based on AIC, BIC), machine learning based selection method (support vector machine recursive feature selection; SVM-REF), and stepwise selection method using the area under the receiver operating characteristic curve (Step-AUC). Based on the simulation and real data analysis, we suggest a Step-AUC method to maximise the prediction performance of the early diagnosis by protein data.
- 1 January 2016
- Published by Inderscience Publishers in International Journal of Data Mining and Bioinformatics
- Vol. 16 (1) , 64
Comparative studies for developing protein based cancer prediction model to maximise the ROC-AUC with various variable selection methods
The era of protein data analysis is coming with more accurate quantification experiments such as the multiple reaction monitoring (MRM). Protein is easier to obtain than the other genetic variants or gene expression data, which makes it more suitable for early diagnosis of cancer. Each patient has unique patterns of protein data, which makes it imperative for the researcher to select the effective markers to construct a consistent model to predict the patients. This research focuses on finding the most effective variable selection method to be applied in the early diagnosis of the pancreatic cancer. In the process, we compare classical selection methods (stepwise selection based on AIC, BIC), machine learning based selection method (support vector machine recursive feature selection; SVM-REF), and stepwise selection method using the area under the receiver operating characteristic curve (Step-AUC). Based on the simulation and real data analysis, we suggest a Step-AUC method to maximise the prediction performance of the early diagnosis by protein data.
- 20 December 2011
- Published by Springer Nature
- Vol. 850, 371-397
Single Marker Family-Based Association Analysis Not Conditional on Parental Information
Family-based association analysis unconditional on parental genotypes models the effects of observed genotypes. This approach has been shown to have greater power than conditional methods. In this chapter, I review two popular association analysis methods accounting for familial correlations: the marginal model using generalized estimating equations (GEE) and the mixed model with a polygenic random component. The marginal approach does not explicitly model familial correlations but uses the information to improve the efficiency of parameter estimates. This model, using GEE, is useful when the correlation structure is not of interest; the correlations are treated as nuisance parameters. In the mixed model, familial correlations are modeled as random effects, e.g., the polygenic inheritance model accounts for correlations originating from shared genomic components within a family. These unconditional methods provide a flexible modeling framework for general pedigree data to accommodate traits with various distributions and many types of covariate effects. The analysis procedures are demonstrated using the ASSOC program in the S.A.G.E. package and the R package gee, including how to prepare input data, conduct the analysis, and interpret the output. ASSOC allows models to include random components of additional familial correlations that may be not sufficiently explained by a polygenic effect and addresses nonnormality of response variables by transformation methods. With its ease of use, ASSOC provides a useful tool for association analysis of large pedigree data.
- 20 December 2011
- Published by Springer Nature
- Vol. 850, 371-397
Single Marker Family-Based Association Analysis Not Conditional on Parental Information
Family-based association analysis unconditional on parental genotypes models the effects of observed genotypes. This approach has been shown to have greater power than conditional methods. In this chapter, I review two popular association analysis methods accounting for familial correlations: the marginal model using generalized estimating equations (GEE) and the mixed model with a polygenic random component. The marginal approach does not explicitly model familial correlations but uses the information to improve the efficiency of parameter estimates. This model, using GEE, is useful when the correlation structure is not of interest; the correlations are treated as nuisance parameters. In the mixed model, familial correlations are modeled as random effects, e.g., the polygenic inheritance model accounts for correlations originating from shared genomic components within a family. These unconditional methods provide a flexible modeling framework for general pedigree data to accommodate traits with various distributions and many types of covariate effects. The analysis procedures are demonstrated using the ASSOC program in the S.A.G.E. package and the R package gee, including how to prepare input data, conduct the analysis, and interpret the output. ASSOC allows models to include random components of additional familial correlations that may be not sufficiently explained by a polygenic effect and addresses nonnormality of response variables by transformation methods. With its ease of use, ASSOC provides a useful tool for association analysis of large pedigree data.
- Open Access
- 15 December 2009
- Published by Springer Nature in BMC Proceedings
- Vol. 3 (S7) , S34
Genome-wide analysis of haplotype interaction for the data from the North American Rheumatoid Arthritis Consortium
Recent genome-wide association studies on several complex diseases have focused on individual single-nucleotide polymorphism (SNP) analysis; however, not many studies have reported interactions among genes perhaps because the gene-gene and gene-environment interaction analysis could be infeasible due to heavy computing requirements. In this study we propose a new strategy for exploring the interactions among haplotypes. The proposed method consists of two steps. Step 1 tests the single-SNP association of whole genome with multiple testing corrections and finds the haplotype blocks of the significant SNPs. Step 2 performs interaction analysis of haplotypes within blocks. Our proposed method is applied to the rheumatoid arthritis data for Genetic Analysis Workshop 16.
- 15 December 2009
- Published by Springer Nature in BMC Proceedings
- Vol. 3 (S7) , S34
Open AccessGenome-wide analysis of haplotype interaction for the data from the North American Rheumatoid Arthritis Consortium
Recent genome-wide association studies on several complex diseases have focused on individual single-nucleotide polymorphism (SNP) analysis; however, not many studies have reported interactions among genes perhaps because the gene-gene and gene-environment interaction analysis could be infeasible due to heavy computing requirements. In this study we propose a new strategy for exploring the interactions among haplotypes. The proposed method consists of two steps. Step 1 tests the single-SNP association of whole genome with multiple testing corrections and finds the haplotype blocks of the significant SNPs. Step 2 performs interaction analysis of haplotypes within blocks. Our proposed method is applied to the rheumatoid arthritis data for Genetic Analysis Workshop 16.
- Open Access
- 18 December 2007
- Published by Springer Nature in BMC Proceedings
- Vol. 1 (S1) , S69
Identification of expression quantitative trait loci by the interaction analysis using genetic algorithm
Many genes with major effects on quantitative traits have been reported to interact with other genes. However, finding a group of interacting genes from thousands of SNPs is challenging. Hence, an efficient and robust algorithm is needed. The genetic algorithm (GA) is useful in searching for the optimal solution from a very large searchable space. In this study, we show that genome-wide interaction analysis using GA and a statistical interaction model can provide a practical method to detect biologically interacting loci. We focus our search on transcriptional regulators by analyzing gene × gene interactions for cancer-related genes. The expression values of three cancer-related genes were selected from the expression data of the Genetic Analysis Workshop 15 Problem 1 data set. We implemented a GA to identify the expression quantitative trait loci that are significantly associated with expression levels of the cancer-related genes. The time complexity of the GA was compared with that of an exhaustive search algorithm. As a result, our GA, which included heuristic methods, such as archive, elitism, and local search, has greatly reduced computational time in a genome-wide search for gene × gene interactions. In general, the GA took one-fifth the computation time of an exhaustive search for the most significant pair of single-nucleotide polymorphisms.
- 18 December 2007
- Published by Springer Nature in BMC Proceedings
- Vol. 1 (S1) , S69
Open AccessIdentification of expression quantitative trait loci by the interaction analysis using genetic algorithm
Many genes with major effects on quantitative traits have been reported to interact with other genes. However, finding a group of interacting genes from thousands of SNPs is challenging. Hence, an efficient and robust algorithm is needed. The genetic algorithm (GA) is useful in searching for the optimal solution from a very large searchable space. In this study, we show that genome-wide interaction analysis using GA and a statistical interaction model can provide a practical method to detect biologically interacting loci. We focus our search on transcriptional regulators by analyzing gene × gene interactions for cancer-related genes. The expression values of three cancer-related genes were selected from the expression data of the Genetic Analysis Workshop 15 Problem 1 data set. We implemented a GA to identify the expression quantitative trait loci that are significantly associated with expression levels of the cancer-related genes. The time complexity of the GA was compared with that of an exhaustive search algorithm. As a result, our GA, which included heuristic methods, such as archive, elitism, and local search, has greatly reduced computational time in a genome-wide search for gene × gene interactions. In general, the GA took one-fifth the computation time of an exhaustive search for the most significant pair of single-nucleotide polymorphisms.