I am a PhD candidate in Computer Engineering at Boğaziçi University, advised by Arzucan Özgür and Elif Özkırımlı.
My research focuses on machine learning for biomolecular modeling, particularly protein representation learning, function prediction, molecular generation, and the characterization of functional units in proteins.
My previous work has also explored chemical representation learning and natural language processing, including Turkish language modeling and hate speech detection.
News
- Jul 2026 — PUFFIN was accepted to the ISMB 2026 and published as conference proceedings in Bioinformatics.
- May 2026 — STAR-GO was selected for an ISMB 2026 Highlights talk.
- May 2026 — ConGen was published in Molecular Informatics.
- Apr 2026 — STAR-GO was accepted to Bioinformatics.
- Nov 2025 — HATECAT-TR was published in Findings of EMNLP 2025.
Selected publications
PUFFIN: Protein Unit Discovery with Functional Supervision 🪶Puffins are remarkable seabirds adapted for both flight and underwater swimming. This dual adaptation offers a loose parallel to our model, which is also jointly optimized for two objectives.
G. Uludoğan, B. Giledereli, E. Özkırımlı, A. Özgür
Bioinformatics · ISMB Proceedings · 2026
STAR-GO: Improving Protein Function Prediction by Learning to Hierarchically Integrate Ontology-Informed Semantic Embeddings
M. E. Akça, G. Uludoğan, A. Özgür, İ. M. Baytaş
Bioinformatics, 42(4), btag146 · 2026
@article{akca2026stargo,
title={STAR-GO: Improving Protein Function Prediction by Learning to Hierarchically Integrate Ontology-Informed Semantic Embeddings},
author={Akça, M. E. and Uludoğan, G. and Özgür, A. and Baytaş, İ. M.},
journal={Bioinformatics}, volume={42}, number={4}, pages={btag146}, year={2026}
}
ConGen: Targeted Molecule Generation Through Contrastive Learning and Latent Optimization
C. Koban, G. Uludoğan, E. Özkırımlı, A. Özgür
Molecular Informatics, 45(5), e70038 · 2026
HATECAT-TR: A Hate Speech Span Detection and Categorization Dataset for Turkish 🐈No cats were hated in this research. I love cat, even with their unmistakable attitude, surprise scratches, and talent for taking over every workspace. HATECAT-TR refers to hate-speech categorization in Turkish.
H. K. Şeker, G. Uludoğan, P. Önal, A. Özgür
Findings of EMNLP · 2025
Overview of the Hate Speech Detection in Turkish and Arabic Tweets (HSD-2Lang) Shared Task at CASE 2024
G. Uludoğan, S. Dehghan, İ. Arın, E. Erol, B. Yanıkoğlu, A. Özgür
Proceedings of CASE, 229–233 · 2024
@inproceedings{uludogan2024hsd2lang,
title={Overview of the Hate Speech Detection in Turkish and Arabic Tweets (HSD-2Lang) Shared Task at CASE 2024},
author={Uludoğan, G. and Dehghan, S. and Arın, İ. and Erol, E. and Yanıkoğlu, B. and Özgür, A.},
booktitle={Proceedings of the 7th Workshop on Challenges and Applications of Automated Extraction of Socio-Political Events from Text},
pages={229--233}, year={2024}
}
TURNA: A Turkish Encoder–Decoder Language Model for Enhanced Understanding and Generation 🪶Turna is the Turkish word for “crane,” a bird with a prominent place in Anatolian stories, songs, and folk culture. We chose the name to give our Turkish language model a distinctive cultural identity.
G. Uludoğan, Z. Y. Balal, F. Akkurt, M. Türker, O. Güngör, S. Üsküdarlı
Findings of ACL, 10103–10117 · 2024
@inproceedings{uludogan2024turna,
title={TURNA: A Turkish Encoder--Decoder Language Model for Enhanced Understanding and Generation},
author={Uludoğan, G. and Balal, Z. Y. and Akkurt, F. and Türker, M. and Güngör, O. and Üsküdarlı, S.},
booktitle={Findings of ACL}, pages={10103--10117}, year={2024}
}
Exploring Data-Driven Chemical SMILES Tokenization Approaches to Identify Key Protein–Ligand Binding Moieties
A. B. Temizer, G. Uludoğan, R. Özçelik, T. Koulani, E. Özkırımlı, K. O. Ülgen, N. L. Karalı, A. Özgür
Molecular Informatics, 43(3), e202300249 · 2024
@article{temizer2024smiles,
title={Exploring Data-Driven Chemical SMILES Tokenization Approaches to Identify Key Protein--Ligand Binding Moieties},
author={Temizer, A. B. and Uludoğan, G. and others},
journal={Molecular Informatics}, volume={43}, number={3}, pages={e202300249}, year={2024}
}
Exploiting Pretrained Biochemical Language Models for Targeted Drug Design
G. Uludoğan, E. Özkırımlı, K. O. Ülgen, N. L. Karalı, A. Özgür
Bioinformatics, 38(Suppl. 2), ii155–ii161 · 2022
@article{uludogan2022biochemical,
title={Exploiting Pretrained Biochemical Language Models for Targeted Drug Design},
author={Uludoğan, G. and Özkırımlı, E. and Ülgen, K. O. and Karalı, N. L. and Özgür, A.},
journal={Bioinformatics}, volume={38}, number={Supplement 2}, pages={ii155--ii161}, year={2022}
}View the full publication record on Google Scholar →
Experience
Nov 2023 — Present
Researcher · Understanding the Language of Life
Boğaziçi University · European Research Council–funded project
Developing machine-learning methods for protein representation and function, including protein-unit discovery, ontology-informed prediction, and the systematic evaluation of protein function prediction models.
Jun 2023 — Mar 2025
Researcher · Digital Technology for Social Cohesion
Boğaziçi University · EuropeAid-funded project
Developed datasets and computational methods for Turkish hate speech detection and contributed to annotation design, multilingual shared-task organization, and related research publications.
Apr 2020 — Oct 2023
Researcher · Chemical Language Processing for Target-Based Drug Design
Boğaziçi University · TÜBİTAK 119E133
Developed language-model-based methods for targeted drug design, including molecular tokenization, protein-conditioned molecule generation, and latent-space optimization.
Teaching
Teaching Assistant · 2019 — Present
Department of Computer Engineering, Boğaziçi University
I deliver problem sessions and labs, design assessments and grading infrastructure, and mentor student projects across undergraduate and graduate courses.
CMPE 150
Introduction to Computing
CMPE 230
Systems Programming
CMPE 321
Introduction to Database Systems
CMPE 343
Introduction to Probability and Statistics
CMPE 344
Computer Organization
CMPE 493
Information Retrieval
CMPE 549
Bioinformatics
This website is adapted from AcadHomepage, with custom content and design for this site.