🎉 Two ISMB/ECCB 2025 Papers Accepted! A Small Win for AI x Drug Discovery 🧪

I’m excited to share that two of our papers were accepted at ISMB/ECCB 2025 🎉, the world’s largest bioinformatics conference, and I’m joining as co-first author on both.

Having two first-author papers accepted at ISMB is rare even globally (CMU Bioinformatics posted about it when it happened in 2018 📰), and I’m grateful to contribute to this milestone with Prof. Sun Kim’s lab at Seoul National University.

📌 Paper 1: “ADME-Drug-Likeness: Enriching Molecular Foundation Models via Pharmacokinetics-Guided Multi-Task Learning for Drug-likeness Prediction.” We integrate ADME data with molecular foundation models using multi-task learning to predict drug-likeness, moving beyond the limitations of old rule-based filters.

📌 Paper 2: “MixingDTA: Improved Drug-Target Affinity Prediction by Extending Mixup with Guilt-By-Association.” We improve drug-target affinity prediction by blending Mixup augmentation with guilt-by-association, tackling data scarcity and enhancing prediction robustness.

Both will be presented on July 24 in Liverpool, and I’m excited to discuss with peers and mentors in the field. 🌐

Beyond the papers themselves, this is part of our broader mission to advance AI for drug discovery, molecular modeling, and biomedical knowledge integration. If you’re a student interested in these areas, I highly encourage you to explore research—bioinformatics is a deeply rewarding field where small models can lead to big biological insights.

Thanks for reading, and I’ll share reflections and slides after the conferences! 🫶