JBJe-Hyun Baek

Je-Hyun Baek

*This is an algorithmically generated author record based on publicly available information.

Je-Hyun Baek has a total of 57 co-authors. Here are the top co-authors by number of shared publications:

  • Won Suk Yang

    R&D Center for Clinical Mass Spectrometry, Seegene Medical Foundation, Seoul, Korea

    8 shared publications
  • Saeyoung Lee

    R&D Center for Clinical Mass Spectrometry, Seegene Medical Foundation, Seoul, Korea

    6 shared publications
  • Seohyun Hwang

    R&D Center for Clinical Mass Spectrometry Seegene Medical Foundation Seoul Republic of Korea

    5 shared publications
  • Hyojin Kim

    R&D Center for Clinical Mass Spectrometry, Seegene Medical Foundation, Seoul, Korea

    4 shared publications
  • Min Jin Kim

    Department of Laboratory Medicine Seegene Medical Foundation Seoul Republic of Korea

    4 shared publications
  • Dong Huey Cheon

    R&D Center for Clinical Mass Spectrometry, Seegene Medical Foundation, Seongdong-gu, Seoul 04805, Korea

    3 shared publications

Latest Publications

  • Open Access

    GAFAD: A liquid chromatography-tandem mass spectrometry-based model for early hepatocellular carcinoma detection beyond GALAD’s limitations

    by Hyojin Kim, Wonseok Oh, Juri Park, Saeyoung Lee, Won Suk Yang, Soon Sun Kim, Jae Youn Cheong, Je-Hyun Baek

    Background/Aims: The GALAD (Gender, Age, Lens culinaris agglutinin-reactive alpha-fetoprotein [AFP-L3], alpha-fetoprotein [AFP], and des-γ-carboxy prothrombin) score, widely used for hepatocellular carcinoma (HCC) detection, was primarily derived from cohorts with advanced-stage tumors and elevated biomarker levels, potentially overestimating accuracy in early-stage disease. Furthermore, the lectin-based AFP-L3 assay has poor sensitivity at low AFP concentrations, limiting detection of small or AFP-negative tumors.Methods: We developed GAFAD, a multivariable model replacing AFP-L3 with fucosylated AFP percentage, quantified by a validated liquid chromatography–tandem mass spectrometry assay. The model was trained and tested using a hepatitis B virus (HBV)-related cohort (HCC n=235; non-HCC n=290), a diagnostically challenging set with substantial overlap in biomarker levels between HCC and non-HCC. Moreover, a final model (GAFAD) was validated in two independent cohorts (HCC n=210; non-HCC n=245), comprising HBV-, HCV-related and non-viral etiologies.Results: In the development cohort, GAFAD showed superior diagnostic performance to GALAD for distinguishing HCC from non-HCC, with a higher area under the receiver operating characteristic curve (AUC, 0.938 vs. 0.887; P<0.0001) and greater sensitivity (82% vs. 66%) and accuracy (86% vs. 79%) at 90% specificity. In the external validation cohort, GAFAD similarly outperformed GALAD, achieving a higher AUC (0.874 vs. 0.841, P<0.05), greater sensitivity (72% vs. 57%), and improved accuracy (82% vs. 75%) at 90% specificity. This superiority extended to early-stage, very-early-stage, and AFP-negative HCC.Conclusions: GAFAD provides a reliable and generalizable tool for early HCC detection across diverse etiologies, supporting its clinical applicability in surveillance and diagnosis.


  • Open Access

    Development and Validation of a Lectin-independent Liquid Chromatography–Tandem Mass Spectrometry Method for Serum Glycosylated Alpha-fetoprotein Analysis and Comparison with a Liquid-phase Binding Assay

    by Hyojin Kim, Juri Park, Hanseul Suh, Saeyoung Lee, Yoonha Park, Won Suk Yang, Dohsik Minn, Soon Sun Kim, Jae Youn Cheong, Je-Hyun Baek

    Hyojin Kim, M.S., Juri Park, M.S., Hanseul Suh, M.S., Saeyoung Lee, Ph.D., Yoonha Park, B.S., Won Suk Yang, Ph.D., Dohsik Minn, M.D., Soon Sun Kim, M.D., Jae Youn Cheong, M.D., and Je-Hyun Baek, Ph.D.. Ann Lab Med -0001;0:. https://doi.org/10.3343/alm.2025.0003


  • Direct Identification of IMP and VIM Metallo‐β‐Lactamases in Gram‐Negative Bacteria Using A‐MALDI

    by Dong Huey Cheon, Seohyun Hwang, Yoon Kyung Choi, Saeyoung Lee, Won Suk Yang, Bo Kyung Kim, Min Jin Kim, Sun Hwa Lee, Je‐Hyun Baek

    Rationale: Rapid and accurate identification of carbapenemase‐producing Enterobacteriaceae (CPE) is crucial for effective infection control and patient treatment. However, accurate identification of VIM and IMP metallo‐β‐lactamases remains still challenging using MALDI‐TOF MS. Methods: The A‐MALDI method which incorporates sequential lysis steps and internal mass calibration, was used for the identification of IMP and VIM carbapenemases. Two Escherichia coli standard strains and 26 clinical isolates harboring IMP or VIM genes were tested by A‐MALDI along with corresponding negative controls. Previously published carbapenemases‐negative data ( n = 112) were used to check the specificity of IMP and VIM identification. Results: For IMP‐positive isolates, proteoforms corresponding to amino acid residues 20–246 showed distinct peaks for IMP. For VIM‐positive isolates, unique single peaks corresponding to amino acid residues 27–266 allowed clear identification of VIM. Clinical evaluation of A‐MALDI demonstrated 93.9% accuracy for IMP identification (100% sensitivity, 93.3% specificity) and 100% accuracy for VIM identification. Conclusions: This study successfully expands the direct identification capabilities for VIM and IMP, achieving comprehensive identification of all six carbapenemases using A‐MALDI. We anticipate that A‐MALDI will provide clinical laboratories with a powerful tool for rapid identification of all types of carbapenemases in bacterial infections.

  • Open Access

    Clinical evaluation of advanced MALDI-TOF MS for carbapenemase subtyping in Gram-negative isolates

    by Dong Huey Cheon, Heejung Jang, Yoon Kyung Choi, Won Seok Oh, Seohyun Hwang, Ju-Ri Park, Hyojin Kim, Yoonha Park, Saeyoung Lee, Won Suk Yang, et al.

    The spread of carbapenemase-producing Enterobacterales (CPE) is emerging as a significant clinical concern in tertiary hospitals and, in particular, long-term care facilities with deficiencies in infection control. This study aims to evaluate an advanced matrix-assisted laser desorption/ionization (A-MALDI) mass spectrometry method for the identification of carbapenemases and further discrimination of their subtypes in clinical isolates. The A-MALDI method was employed to detect CPE target proteins. Enhancements were made to improve detectability and mass accuracy through the optimization of MALDI-TOF settings and internal mass calibration. A total of 581 clinical isolates were analyzed, including 469 CPE isolates (388 Klebsiella pneumoniae carbapenemases [KPC], 51 NDM, 40 OXA, and 2 GES) and 112 carbapenemase-negative isolates. Clinical evaluation of the A-MALDI demonstrated 100% accuracy and precision in identifying all the collected CPE isolates. Additionally, A-MALDI successfully discriminated individual carbapenemase subtypes (KPC-2 or KPC-3/KPC-4, OXA-48 or OXA-181 or OXA-232, GES-5 or GES-24) and also differentiated co-producing carbapenemase strains (KPC and NDM, KPC and OXA, KPC and GES, and NDM and OXA), attributed to its high mass accuracy and simultaneous detection capability. A-MALDI is considered a valuable diagnostic tool for accurately identifying CPE and carbapenemase’s subtypes in clinical isolates. It may also aid in selecting appropriate antibiotics for each carbapenemase subtype. Ultimately, we expect that the A-MALDI method will contribute to preventing the spread of antibiotic resistance and improving human public health. IMPORTANCE: A-MALDI clearly demonstrated excellent ability to identify CPEs such as KPC, NDM, OXA, and GES when carbapenemase is present in the strain (100% accuracy and precision). The method also successfully discriminated carbapenemase subtypes and simultaneous detection of co-producing multiple carbapenemases in a single strain. This is the first report for simultaneous and multiple detection of intact carbapenemases of KPC, NDM, OXA, and GES using matrix-assisted laser desorption/ionization mass spectrometry in a clinical isolate.


  • Improving mass accuracy in matrix‐assisted laser desorption ionization time‐of‐flight mass spectrometry analysis of pathogenic proteins using 6xHIS‐tagged internal calibration

    by Saeyoung Lee, Ju‐Ri Park, Seohyun Hwang, Won Suk Yang, Je‐Hyun Baek

    Rationale: Linear mode of matrix‐assisted laser desorption ionization time‐of‐flight mass spectrometry (MALDI‐TOFMS) has been routinely used for bacterial identification in the clinic, depending on the pattern analysis of spectral libraries rather than accurate mass measurement of ribosomal proteins (10–15 kDa). However, a demand for more accurate mass analysis of pathogens (e.g. KPC‐2 carbapenemase) has been recently increasing for diagnostic purposes. Methods: We introduced a 6xHIS‐tagged KPC‐2 (i.e. hKPC‐2) and used it as an internal mass calibrator for the mass calibration of target proteins. After internal mass calibration (In‐Cal), we evaluated the observed mass of KPC‐2 against the theoretical mass of hKPC‐2, which has 823 Da mass difference from the target protein. We further assessed the accuracy and precision of our calibration method regarding the identification of KPC‐2 and other pathogens in clinical isolates ( n = 42). Results: Among several candidates for internal mass calibrators, the In‐Cal using a 6xHIS‐tagged protein on the target showed the highest mass accuracy and precision in the detection of target proteins (e.g. KPC‐2). The application of hKPC‐2 as an internal calibrator showed substantial improvement of mass accuracy, mass precision and also quantification of KPC in linearity and repeatability for KPC detection in the clinical isolates. Conclusions: Our In‐Cal method using 6xHIS‐tagged protein in MALDI‐TOFMS allows successful mass calibration (20 kDa) and provides high mass accuracy as much as that of medium‐ and high‐resolution mass spectrometry.


  • A Graphene-Coated Silicon Wafer Plate Improves the Sensitivity and Reproducibility of MALDI-TOF MS Analysis of Proteins

    by Yoon Kyung Choi, Dong Huey Cheon, Won Suk Yang, Je-Hyun Baek

    Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) is widely used to analyze small and large molecules. However, proteins are difficult to analyze with MALDI-TOF MS in clinical applications because of their low ionization efficiency and heterogeneous crystallization with the matrix on the sample spots. Here, we investigate the potential of a customized graphene-coated silicon wafer (G/SiO2) plate for MALDI-TOF MS analysis of a clinically important protein, KPC-2, in comparison with a conventional stainless steel (SUS) plate. Our results demonstrate that the G/SiO2 plate outperforms the SUS plate in terms of sensitivity, reproducibility, and mass accuracy/precision across a wide range of molecular weights, even with highly complex samples. Furthermore, a five-day robustness test confirms the practical applicability of the G/SiO2 plate for the reliable identification of target protein(s) in MALDI-TOF MS analysis. Overall, our findings suggest that the use of the G/SiO2 plate holds great potential for improving the sensitivity and reproducibility of MALDI-TOF MS analysis for the identification of proteins, making it a promising tool for clinical applications.


  • Fast and straightforward simultaneous quantification of multiple apolipoproteins in human serum on a high‐throughput LC‐MS/MS platform

    by Hyojin Kim, Won Suk Yang, Dongheui An, Sang‐Guk Lee, Je‐Hyun Baek

    Purpose: Apolipoprotein monitoring is useful for diagnosing cardiovascular diseases, as they are risk factors of arteriosclerosis and other neutral fat‐related diseases. Liquid chromatography‐tandem mass spectrometry (LC‐MS/MS) is advantageous for simultaneous apolipoprotein quantification, differentiation, and standardization including their isoforms. However, fast and straightforward sample preparation that retains quantification accuracy remains challenging in clinical MS.Experimental Design: We developed a simultaneous assay for serum apolipoprotein A‐I (ApoA‐I), apolipoprotein B100 family, and apolipoprotein C‐III (ApoC‐III) using a high‐throughput LC‐MS/MS platform coupled with a BRAVO system. The assay was simplified by using sodium deoxycholate and trypsin/lys‐C without reduction and alkylation steps.Results: Simple sample preparation reduced turnaround time by 1.5 h and neat goat serum was chosen as an optimal calibration matrix for accurate protein quantification. Assay precision, linearity, correlation, accuracy, limit of detection (LOD), limit of quantitation (LOQ), and carryover were validated according to CLSI guidelines over 41 days using more than 100 human serum samples. Good correlation compared with turbidimetric immunoassay (TIA) was observed by Deming regression for all analytes.Conclusions and Clinical Relevance: A high‐throughput LC‐MS/MS and BRAVO assay for simultaneous apolipoprotein analysis was validated using a simple preparation method with a human serum calibrator in goat serum matrix. The assay is readily expandable to include other target serum proteins and/or their isoforms.


  • Simultaneous Quantification of Apolipoprotein C-III O-Glycoforms by Protein-MRM

    by Hyojin Kim, Dong Huey Cheon, Won Suk Yang, Je-Hyun Baek

    Apolipoprotein C-III (APOC-III) regulates triglyceride levels, associated with a risk of cardiovascular disease. One gene generates several proteoforms, each with a different molecular mass and a unique function. Unlike peptide multiple reaction monitoring (MRM), protein-MRM without digestion is required to analyze clinically relevant individual proteoforms. We developed a protein-MRM method without digestion to individually quantify APOC-III proteoforms in human serum. We optimized the protein-MRM method following 60% acetonitrile extraction with C18 filtration. Bovine serum and myoglobin served as supporting cushions and the internal standard during sample preparation, respectively. Furthermore, we evaluated the LOD, lower limit of quantification, linearity, accuracy, and precision. Good correlation compared with turbidimetric immunoassay (TIA) and peptide-MRM was observed using 30 clinical sera. Individual APOC-III O-glycoforms were identified by top-down proteomics and simultaneously quantified using the protein-MRM method. The sum abundance of APOC-III proteoforms was significantly correlated with TIA and peptide-MRM. Our protein-MRM method provides an affordable and rapid quantification of potential disease-specific proteoforms. Precise quantification of each proteoform allows investigators to identify novel biological roles potentially related to cardiovascular disease or novel biomarkers. We expect our protein-oriented method to be more clinically useful than antibody-based immunoassays and peptide-oriented MRM analysis, especially for quantification of a biomarker proteoform with certain post-translational modifications.


  • Clinical Performance of the Osmotic Shock-MALDI MS Method to Detect Klebsiella pneumoniae Carbapenemase in Clinical Isolates

    by Saeyoung Lee, Seohyun Hwang, Dong Huey Cheon, Heejung Jang, Won Suk Yang, Min Jin Kim, SunHwa Lee, Je-Hyun Baek

    The World Health Organization recently highlighted the serious worldwide problem of the emergence of antibiotic-resistant or antibiotic multidrug-resistant bacteria. Carbapenem-resistant Enterobacterales, including carbapenemase-producing Enterobacterales (CPE), are major antibiotic-resistant bacteria that can be identified by various methods, including antibiotic susceptibility testing, PCR, and immunologic assays.


  • Nutrient‐specific proteomic analysis of the mucin degrading bacterium Akkermansia muciniphila

    by Ji‐Young Lee, Hyeon‐Su Jin, Kyoung Su Kim, Je‐Hyun Baek, Bong‐Soo Kim, Dong‐Woo Lee

    Akkermansia muciniphila is a prominent mucin‐degrading bacterium that acts as a keystone species in regulating the human gut microbiota. Despite recently increasing research into this bacterium and its relevance to human health, a high‐resolution database of its functional proteins remains scarce. Here, we provide a proteomic overview of A. muciniphila grown in different nutrient conditions ranging from defined to complex. Of 2318 protein‐coding genes in the genome, we identified 841 (40%) that were expressed at the protein level. Overall, proteins involved in energy production and carbohydrate metabolism indicate that A. muciniphila relies mainly on the Embden‐Meyerhof‐Parnas pathway, and produces short‐chain fatty acids through anaerobic fermentation in a nutrient‐specific manner. Moreover, this bacterium possesses a broad repertoire of glycosyl hydrolases, together with putative peptidases and sulfatases, to cleave O‐glycosylated mucin. Of them, putative mucin‐degrading enzymes (Amuc_1220, Amuc_1120, Amuc_0052, Amuc_0480, and Amuc_0060) are highly abundant in the mucin‐supplemented media. Furthermore, A. muciniphila uses mucin‐derived monosaccharides as sources of energy and cell wall biogenesis. Our dataset provides nutrient‐dependent global proteomes of A. muciniphila ATCC BAA‐835 to offer insights into its metabolic functions that shape the composition of the human gut microbiota via mucin degradation.