Hyun-Seok Oh
Hyun-Seok Oh has a total of 59 co-authors. Here are the top co-authors by number of shared publications:
Hyun Kim
CJ Bioscience, Inc., Seoul, Republic of Korea
2 shared publications- Jongsik Chun
Interdisciplinary Program in Bioinformatics, Seoul National University, Seoul, Republic of Korea
2 shared publications - Kihyun Lee
CJ Bioscience Inc, Seoul, 04527, Republic of Korea
2 shared publications Mauricio Chalita
CJ Bioscience, Inc., Seoul, Republic of Korea
2 shared publications- Sein Park
Interdisciplinary Program in Bioinformatics, Seoul National University, Seoul, 08826, Republic of Korea
2 shared publications - Yeonjae Jung
Interdisciplinary Program in Bioinformatics, Seoul National University, Seoul, 08826, Republic of Korea
2 shared publications
Latest Publications
- Open Access
- 6 August 2026
- Published by Springer Nature in Genome Medicine
- Vol. 18 (1)
Integrative stratification of response to immunotherapy using host and microbial signatures in patients with advanced non-small cell lung cancer
Background: The gut microbiome has been linked to responses to immune checkpoint inhibitors (ICIs), yet longitudinal multi-omics studies remain limited. Methods: Here, we serially collected gut microbiome data and performed multi-omics analyses, including RNA sequencing (RNA-seq) and whole-exome sequencing (WES), on samples from 92 patients with advanced non-small cell lung cancer (NSCLC) treated with pembrolizumab, to identify factors contributing to treatment response. We compared gut microbiome data with those of a healthy cohort ( n = 92) and examined changes in gut microbial signatures during immunotherapy. Gut microbiome data from publicly available ICI-treated cohorts ( n = 728) were also used together with our own cohort to develop and validate a score for classifying response to ICIs. Results: Responders (defined as complete response, partial response, or stable disease of at least 6 months) were enriched for Faecalibacterium , while non-responders were enriched for Lactobacillus and Escherichia , with these taxa showing specific associations with dietary fiber intake and antibiotic use. Dynamic shifts in dysbiotic signatures were observed throughout treatment and disease progression, with dissimilarity from the healthy cohort decreasing after treatment only in responders but increasing at progression in both responders and non-responders. We also developed a cross-population score based on the ratio of responder- to non-responder-associated microbial markers. This score showed a significant positive correlation with the index reflecting gut microbial health status and improved response stratification when combined with tumor mutational burden (TMB). Conclusions: These findings highlight the importance of characterizing dysbiosis using gut microbial signatures, which act along a distinct and complementary axis to explain responses to ICIs.
- 6 August 2026
- Published by Springer Nature in Genome Medicine
- Vol. 18 (1)
Open AccessIntegrative stratification of response to immunotherapy using host and microbial signatures in patients with advanced non-small cell lung cancer
Background: The gut microbiome has been linked to responses to immune checkpoint inhibitors (ICIs), yet longitudinal multi-omics studies remain limited. Methods: Here, we serially collected gut microbiome data and performed multi-omics analyses, including RNA sequencing (RNA-seq) and whole-exome sequencing (WES), on samples from 92 patients with advanced non-small cell lung cancer (NSCLC) treated with pembrolizumab, to identify factors contributing to treatment response. We compared gut microbiome data with those of a healthy cohort ( n = 92) and examined changes in gut microbial signatures during immunotherapy. Gut microbiome data from publicly available ICI-treated cohorts ( n = 728) were also used together with our own cohort to develop and validate a score for classifying response to ICIs. Results: Responders (defined as complete response, partial response, or stable disease of at least 6 months) were enriched for Faecalibacterium , while non-responders were enriched for Lactobacillus and Escherichia , with these taxa showing specific associations with dietary fiber intake and antibiotic use. Dynamic shifts in dysbiotic signatures were observed throughout treatment and disease progression, with dissimilarity from the healthy cohort decreasing after treatment only in responders but increasing at progression in both responders and non-responders. We also developed a cross-population score based on the ratio of responder- to non-responder-associated microbial markers. This score showed a significant positive correlation with the index reflecting gut microbial health status and improved response stratification when combined with tumor mutational burden (TMB). Conclusions: These findings highlight the importance of characterizing dysbiosis using gut microbial signatures, which act along a distinct and complementary axis to explain responses to ICIs.
- Open Access
- 27 February 2026
- Published by American Society of Hematology in Blood Advances
- Vol. 10 (5) , 1634-1645
Microbiome and metabolite biomarkers of CAR T-cell therapy outcomes in relapsed/refractory diffuse large B-cell lymphoma
CD19 chimeric antigen receptor (CAR) T-cell therapy has revolutionized treatment for relapsed/refractory diffuse large B-cell lymphoma (R/R-DLBCL), but challenges such as posttreatment failure and immune-related adverse events (AEs) persist. This study explores the gut microbiome as a predictive biomarker for CAR T-cell therapy outcomes and toxicity. Stool and serum samples from patients with R/R-DLBCL were analyzed at apheresis (47 samples) and 1 month after infusion (32 samples) using whole-genome sequencing metagenomics. When compared with healthy controls and newly-diagnosed DLBCL, R/R-DLBCL showed significant gut dysbiosis, characterized by increased Proteobacteria and Enterobacteriaceae. Responders had higher levels of Bacteroides fragilis, whereas nonresponders exhibited higher levels of Faecalibacterium prausnitzii. Functional metagenomic analysis suggested enrichment of inosine biosynthesis pathways in responders, and elevated serum inosine demonstrated an exploratory association with improved progression-free survival. Distinct microbial taxa and serum fatty acid profiles were also linked to CAR T-cell–related AEs, with higher acetate and butyrate levels in patients without AEs and increased isovalerate in those with AEs. These findings indicate that gut microbiome features—particularly Bacteroides fragilis and inosine metabolism—may serve as candidate biomarkers for CAR T-cell therapy outcomes and toxicity. However, given the exploratory nature of these analyses and the limited cohort size, results should be interpreted cautiously. Larger, prospective studies will be required to validate these observations and to assess the potential of microbiome-based strategies to optimize CAR T-cell therapy in R/R-DLBCL.
- 27 February 2026
- Published by American Society of Hematology in Blood Advances
- Vol. 10 (5) , 1634-1645
Open AccessMicrobiome and metabolite biomarkers of CAR T-cell therapy outcomes in relapsed/refractory diffuse large B-cell lymphoma
CD19 chimeric antigen receptor (CAR) T-cell therapy has revolutionized treatment for relapsed/refractory diffuse large B-cell lymphoma (R/R-DLBCL), but challenges such as posttreatment failure and immune-related adverse events (AEs) persist. This study explores the gut microbiome as a predictive biomarker for CAR T-cell therapy outcomes and toxicity. Stool and serum samples from patients with R/R-DLBCL were analyzed at apheresis (47 samples) and 1 month after infusion (32 samples) using whole-genome sequencing metagenomics. When compared with healthy controls and newly-diagnosed DLBCL, R/R-DLBCL showed significant gut dysbiosis, characterized by increased Proteobacteria and Enterobacteriaceae. Responders had higher levels of Bacteroides fragilis, whereas nonresponders exhibited higher levels of Faecalibacterium prausnitzii. Functional metagenomic analysis suggested enrichment of inosine biosynthesis pathways in responders, and elevated serum inosine demonstrated an exploratory association with improved progression-free survival. Distinct microbial taxa and serum fatty acid profiles were also linked to CAR T-cell–related AEs, with higher acetate and butyrate levels in patients without AEs and increased isovalerate in those with AEs. These findings indicate that gut microbiome features—particularly Bacteroides fragilis and inosine metabolism—may serve as candidate biomarkers for CAR T-cell therapy outcomes and toxicity. However, given the exploratory nature of these analyses and the limited cohort size, results should be interpreted cautiously. Larger, prospective studies will be required to validate these observations and to assess the potential of microbiome-based strategies to optimize CAR T-cell therapy in R/R-DLBCL.
- Open Access
Genome-Level Hierarchical Attention Transformer with Multi-Head Attention Weighted Sum for Broad-Spectrum Antimicrobial Resistance Prediction and Discovery of Resistance-Related Genomic Contexts
Antimicrobial resistance is a growing global health concern, requiring reliable tools for predicting resistance across a wide range of bacteria and antibiotics. In this study, we introduce a genome-level hierarchical attention transformer (GL-HAT) that integrates a pretrained genomic foundation model with hierarchical attention mechanisms to analyze the full protein sequence context of bacterial genomes. GL-HAT is designed to deliver both accurate resistance predictions and interpretable insights into the genomic features associated with resistance using specialized modules. Trained and validated on a comprehensive dataset of bacterial genomes and antimicrobial susceptibility profiles, the model outperformed traditional machine learning method, with an F1-score of 0.845 and AUROC of 0.953. Moreover, GL-HAT could demonstrate the resistance-related contexts in the genome, such as vanA operon and nearby resolvase genes in vancomycin-resistant Enterococcus and tet(M) gene adjacent to conjugation proteins in tetracycline-resistant Streptococcus pneumoniae. By combining high accuracy with interpretability, GL-HAT offers a practical approach for advancing AMR diagnostics and research, supporting more informed clinical and public health decisions.
Open AccessGenome-Level Hierarchical Attention Transformer with Multi-Head Attention Weighted Sum for Broad-Spectrum Antimicrobial Resistance Prediction and Discovery of Resistance-Related Genomic Contexts
Antimicrobial resistance is a growing global health concern, requiring reliable tools for predicting resistance across a wide range of bacteria and antibiotics. In this study, we introduce a genome-level hierarchical attention transformer (GL-HAT) that integrates a pretrained genomic foundation model with hierarchical attention mechanisms to analyze the full protein sequence context of bacterial genomes. GL-HAT is designed to deliver both accurate resistance predictions and interpretable insights into the genomic features associated with resistance using specialized modules. Trained and validated on a comprehensive dataset of bacterial genomes and antimicrobial susceptibility profiles, the model outperformed traditional machine learning method, with an F1-score of 0.845 and AUROC of 0.953. Moreover, GL-HAT could demonstrate the resistance-related contexts in the genome, such as vanA operon and nearby resolvase genes in vancomycin-resistant Enterococcus and tet(M) gene adjacent to conjugation proteins in tetracycline-resistant Streptococcus pneumoniae. By combining high accuracy with interpretability, GL-HAT offers a practical approach for advancing AMR diagnostics and research, supporting more informed clinical and public health decisions.
- Top Cited PapersOpen Access
- 18 June 2024
- Published by Microbiology Society in International Journal of Systematic and Evolutionary Microbiology
- Vol. 74 (6) , 006421
EzBioCloud: a genome-driven database and platform for microbiome identification and discovery
With the continued evolution of DNA sequencing technologies, the role of genome sequence data has become more integral in the classification and identification of Bacteria and Archaea. Six years after introducing EzBioCloud, an integrated platform representing the taxonomic hierarchy of Bacteria and Archaea through quality-controlled 16S rRNA gene and genome sequences, we present an updated version, that further refines and expands its capabilities. The current update recognizes the growing need for accurate taxonomic information as defining a species increasingly relies on genome sequence comparisons. We also incorporated an advanced strategy for addressing underrepresented or less studied lineages, bolstering the comprehensiveness and accuracy of our database. Our rigorous quality control protocols remain, where whole-genome assemblies from the NCBI Assembly Database undergo stringent screening to remove low-quality sequence data. These are then passed through our enhanced identification bioinformatics pipeline which initiates a 16S rRNA gene similarity search and then calculates the average nucleotide identity (ANI). For genome sequences lacking a 16S rRNA sequence and without a closely related genomic representative for ANI calculation, we apply a different ANI approach using bacterial core genes for improved taxonomic placement (core gene ANI, cgANI). Because of the increase in genome sequences available in NCBI and our newly introduced cgANI method, EzBioCloud now encompasses a total of 109 835 species, of which 21 964 have validly published names. 47 896 are candidate species identified either through 16S rRNA sequence similarity (phylotypes) or through whole genome ANI (genomospecies), and the remaining 39 975 were positioned in the taxonomic tree by cgANI (species clusters). Our EzBioCloud database is accessible at www.ezbiocloud.net/db.
- 18 June 2024
- Published by Microbiology Society in International Journal of Systematic and Evolutionary Microbiology
- Vol. 74 (6) , 006421
Top Cited PapersOpen AccessEzBioCloud: a genome-driven database and platform for microbiome identification and discovery
With the continued evolution of DNA sequencing technologies, the role of genome sequence data has become more integral in the classification and identification of Bacteria and Archaea. Six years after introducing EzBioCloud, an integrated platform representing the taxonomic hierarchy of Bacteria and Archaea through quality-controlled 16S rRNA gene and genome sequences, we present an updated version, that further refines and expands its capabilities. The current update recognizes the growing need for accurate taxonomic information as defining a species increasingly relies on genome sequence comparisons. We also incorporated an advanced strategy for addressing underrepresented or less studied lineages, bolstering the comprehensiveness and accuracy of our database. Our rigorous quality control protocols remain, where whole-genome assemblies from the NCBI Assembly Database undergo stringent screening to remove low-quality sequence data. These are then passed through our enhanced identification bioinformatics pipeline which initiates a 16S rRNA gene similarity search and then calculates the average nucleotide identity (ANI). For genome sequences lacking a 16S rRNA sequence and without a closely related genomic representative for ANI calculation, we apply a different ANI approach using bacterial core genes for improved taxonomic placement (core gene ANI, cgANI). Because of the increase in genome sequences available in NCBI and our newly introduced cgANI method, EzBioCloud now encompasses a total of 109 835 species, of which 21 964 have validly published names. 47 896 are candidate species identified either through 16S rRNA sequence similarity (phylotypes) or through whole genome ANI (genomospecies), and the remaining 39 975 were positioned in the taxonomic tree by cgANI (species clusters). Our EzBioCloud database is accessible at www.ezbiocloud.net/db.
- 14 February 2024
- Published by Elsevier in Psychiatry Research
- Vol. 335, 115775
Gut microbial and clinical characteristics of individuals with autism spectrum disorder differ depending on the ecological structure of the gut microbiome
- 14 February 2024
- Published by Elsevier in Psychiatry Research
- Vol. 335, 115775
Gut microbial and clinical characteristics of individuals with autism spectrum disorder differ depending on the ecological structure of the gut microbiome
- Open Access
- 12 January 2023
- Published by Springer Nature in Scientific Reports
- Vol. 13 (1) , 1-11
Altered gut microbiota in individuals with episodic and chronic migraine
Emerging evidence reveals a close association between gut microbiota and human neurological disorders. The present study aimed to assess whether the composition of gut microbiota in participants with episodic migraine (EM) and chronic migraine (CM) was altered in comparison to that of the controls. This study was a cross-sectional, case–control study. The gut microbiota were evaluated by the partial, targeted sequencing of the 16S rRNA V3–V4 region. This study enrolled 42 and 45 participants with EM and CM, respectively, and 43 controls. Alpha and beta diversities revealed no significant difference among the three groups; however, the microbiota composition at the class, order, family, and genus levels differed significantly between EM and the control, CM and the control, and the EM and CM groups. Moreover, higher composition of PAC000195_g was significantly associated with a lower headache frequency among the five genera that exhibited significantly different microbiota composition in EM and CM. Agathobacter revealed a significant negative association with severe headache intensity. The findings of the present study provide evidence of altered gut microbiota in EM and CM. These findings will help in understanding the course and treatment of migraine.
- 12 January 2023
- Published by Springer Nature in Scientific Reports
- Vol. 13 (1) , 1-11
Open AccessAltered gut microbiota in individuals with episodic and chronic migraine
Emerging evidence reveals a close association between gut microbiota and human neurological disorders. The present study aimed to assess whether the composition of gut microbiota in participants with episodic migraine (EM) and chronic migraine (CM) was altered in comparison to that of the controls. This study was a cross-sectional, case–control study. The gut microbiota were evaluated by the partial, targeted sequencing of the 16S rRNA V3–V4 region. This study enrolled 42 and 45 participants with EM and CM, respectively, and 43 controls. Alpha and beta diversities revealed no significant difference among the three groups; however, the microbiota composition at the class, order, family, and genus levels differed significantly between EM and the control, CM and the control, and the EM and CM groups. Moreover, higher composition of PAC000195_g was significantly associated with a lower headache frequency among the five genera that exhibited significantly different microbiota composition in EM and CM. Agathobacter revealed a significant negative association with severe headache intensity. The findings of the present study provide evidence of altered gut microbiota in EM and CM. These findings will help in understanding the course and treatment of migraine.