cMolGPT - A Conditional Generative Pre-Trained Transformer for Molecular Design

---

cMolGPT is a conditional generative pre-trained Transformer for target-aware molecular generation.
By conditioning on protein targets, cMolGPT generates chemically valid, diverse, and biologically relevant molecules tailored to specific therapeutic contexts.

Key features:

• Target-specific de novo molecule generation
• High chemical validity and structural diversity
• Improved hit discovery efficiency
• Generalizable across multiple drug targets

📄 Paper
cMolGPT: A Conditional Generative Pre-Trained Transformer for Target-Specific De Novo Molecular Generation, Molecules, 2023
👉 Read the article (MDPI)

💻 Code
👉 GitHub Repo

---

The model has been evaluated on multiple therapeutic targets (e.g., EGFR, HTR1A, S1PR1), demonstrating its ability to generate target-enriched molecular libraries suitable for downstream screening and lead optimization.

---