ProtBiC: Bipartite context-supervised protein representation learning for function prediction
Manuscript · In revision
Ph.D. Candidate Bioinformatics
Yejin Kan
I develop machine learning methods for representing proteins and predicting their functions.
My work focuses on scalable representation learning and on using functional annotations and biological relationships to guide model training.
Research
Machine learning × computational biology
Representation learning
Learning protein representations from sequence and structural information, guided by biological supervision.
Multi-label learning
Studying how functional annotations and relationships among labels can inform machine learning models.
Protein function prediction
Developing and evaluating methods for predicting protein functions across diverse annotation tasks.
Education
Ph.D. in Computer Science and Artificial Intelligence
Dongguk University, Seoul, South Korea · Advisor: Prof. Gangman Yi
Expected graduation: Aug. 2027
M.S. in Multimedia Engineering
Dongguk University, Seoul, South Korea · Advisor: Prof. Gangman Yi
B.S. in Multimedia Engineering
Dongguk University, Seoul, South Korea
Publications
Manuscripts, journal articles, and conference papers.
Manuscripts
3 manuscriptsResearch in progress; not yet published.
ProtCon: Hierarchy-aware multi-modal contrastive representation learning for protein function prediction
Manuscript · Under review
Crime-risk-based patrol route optimization for crime prevention in urban road networks
Manuscript · In revision
Journal articles
8 papersProtDML: Label-aware representation learning for broad-spectrum protein function prediction
Briefings in Bioinformatics, 27(3), bbag293 (2026)
A Method for Consensus Anchor Graph Learning for Incomplete Multi-View Clustering
IEEE Access, 14, 18155–18168 (2026)
GapSense: Similarity estimation-based gap filler with TGS-reads for genome assemblies
Interdisciplinary Sciences: Computational Life Sciences, 18(2), 501–514 (2026)
DeepPI: Alignment-free analysis of flexible length proteins based on deep learning and image generator
Interdisciplinary Sciences: Computational Life Sciences, 16(3) (2024)
AutoSCAN: Automatic detection of DBSCAN parameters and efficient clustering of data in overlapping density regions
PeerJ Computer Science, 10, e1921 (2024)
Parameter-Free Basis Allocation for Efficient Multiple Metric Learning
Machine Learning: Science and Technology, 4(4), 045049 (2023)
Hotspots-based patrol route optimization algorithm for smart policing
Heliyon, 9(10), e20931 (2023)
cPlot: Contig-Plotting Visualization for the Analysis of Short-Read Nucleotide Sequence Alignments
International Journal of Molecular Sciences, 23(19), 11484 (2022)
Conference papers
3 papersT-MDML: Triplet-based multiple distance metric learning for multi-instance multi-label classification with label correlation
Proceedings of the AAAI Conference on Artificial Intelligence, 39(17), 17850–17858 (2025)
A Novel Gap-Filling Method Based on Hybrid Read Information Analysis
IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 3827–3829 (2022)
Novel Patrol Route Optimization Method Based on Big Data Analysis
IEEE International Conference on Big Data, 6699–6701 (2022)
Conference papers (Korea)
3 papersHS Code Classification Using Graph RAG with an HS Explanatory Notes-Based Knowledge Graph
Proceedings of the Korea Computer Congress (KCC) (2026) · In Korean
Public Data-Based Hotspot Detection and Patrol Route Optimization with Spatiotemporal Data
Proceedings of the Korea Computer Congress (KCC), 414–416 (2024) · In Korean
BLAST-Based Gap Analysis for Gap Filling Algorithm
Proceedings of the Korea Computer Congress (KCC), 1625–1627 (2022) · In Korean
Conference posters and presentations, teaching experience, and other academic activities are listed in my CV (PDF) ↗.
Selected highlights
Additional details are available in my CV.
2022
Principal Investigator · WISET-funded research on a hybrid read-based gap-filling algorithm
2023–2024
Dongguk Global Fellowship · Dongguk University
2021–2022,
2024–2026
2024–2026
Teaching Assistant · Undergraduate computer science courses at Dongguk University
Contact
For research inquiries or collaborations, feel free to get in touch.