Hello, visitor!

This is Rıza! I am a PhD candidate Molecular Machine Learning Group led by Francesca Grisoni. I work on generative deep learning for molecule design and I collaborate extensively with natural scientists. I am particularly interested in (chemical) language modeling and its applications in drug discovery.

I am also a big sports fan 🏀 ⚽ 🟡 🔵 and a food enthusiast 🍲 🍖 🍵 Want to chat? Find me here!

Interests

  • Molecule design
  • Generative deep learning
  • Language modeling

Education

  • Molecular Machine Learning, 2026

    Thesis: Advancing Chemical Language Models for De Novo Drug Design.

  • MSc. in Computer Engineering, 2022

    GPA: 3.94/4.00. Thesis: Biomolecular Language Processing for Drug - Target Affinity Prediction.

  • BSc. in Computer Engineering, 2018

    GPA: 3.57/4.00. Thesis: Disease Classification based on Genomic Data with Machine Learning.

Publications

peptidy: A light-weight Python library for peptide representation in machine learning

We present a python library to encode peptides for machine learning applications. Non-canonical amino acids and post-translational modifications are supported.
Rıza Özçelik, Laura van Weesep, Sarah de Ruiter, Francesca Grisoni
PDF Code DOI
peptidy: A light-weight Python library for peptide representation in machine learning

The Jungle of Generative Drug Discovery: Traps, Treasures, and Ways Out

Surprising pitfalls in common evaluation approaches for molecule libraries generated by deep learning models. Simple solutions are proposed.
Rıza Özçelik, Francesca Grisoni
PDF Code DOI
The Jungle of Generative Drug Discovery: Traps, Treasures, and Ways Out

A Hitchhiker's Guide to Deep Chemical Language Processing for Bioactivity Prediction

Practical guidelines for training deep learning models on molecular string representations for bioactivity prediction.
Rıza Özçelik, Francesca Grisoni
PDF Code DOI
A Hitchhiker's Guide to Deep Chemical Language Processing for Bioactivity Prediction

Deep Supramolecular Language Processing for Co-crystal Prediction

DeepCocrystal is a convolutional neural network to predict co-crystal formation. SMILES augmentation is key to its development.
Rebecca Birolo, Rıza Özçelik, Andrea Aramini, Roberto Gobetto, Michele Remo Chierotti, Francesca Grisoni
PDF DOI
Deep Supramolecular Language Processing for Co-crystal Prediction

Chemical Language Modeling with Structured State Space Sequence Models

A novel approach to chemical language modeling. First application of structured state space sequence models (S4) to de novo design.
Rıza Özçelik, Sarah de Ruiter, Emanuele Crusciolo, Francesca Grisoni
PDF Code
Chemical Language Modeling with Structured State Space Sequence Models
See all publications

Contact 💭

  • r.ozcelik@tue.nl
  • De Zaale, Eindhoven, Noord-Brabant. 5612 AZ
  • Institute of Complex Molecular Systems, Ceres.
  • DM Me