State-of-the-art generative model for crystal structure prediction and de novo generation of inorganic crystals.

DENOPTIM is a software package for de novo design and virtual screening of functional molecules of any kind.

Classification models for hemolytic nature and hemolytic activity predictions in peptide/protein sequences

QBMG: Quasi-Biogenic Molecule Generator with Deep Recurrent Network

Evaluation Framework for AI-driven Molecular Design of Multi-target Drugs

Data, scripts, and software tools for automated evaluation of Ru-based catalysts for olefin metathesis.

Autonomous AI agent for de novo drug design orchestrating generative mutation, ADME/Tox prediction, and 3D Vina/GNINA docking.

Protein Design Agentic System — LLM as CPU, Agents as System, Skills as Software, Toolkits as Pluggable Scientific Capability Packs.

AI-driven de novo drug discovery: LLM-in-the-loop molecular design + Vina docking + MM-GBSA + MD on EGFR T790M/L858R. Runs on a laptop, no commercial software.

A PyTorch-based Character-Level Variational Autoencoder (VAE) development pipeline designed for de novo molecular generation and generative drug design, featuring RDKit validation for novel chemical structures

AI-Driven De Novo Peptide Generator. Designing novel, highly diverse amino acid sequences for targeted biomedical applications.