Boltz2 Notebook – A streamlined Colab-based pipeline for protein structure prediction and binding affinity analysis using the Boltz2 deep learning model.

Python drug discovery toolkit for Boltz2 structure and affinity prediction, BoltzGen binder and antibody design, and TxGemma ADMET integration in one API.

Uses DAP (Distributed Axial Parallelism) to prevent OOM when running Boltz-2 protein structure inference. Optional FlexAttention for triangle attention.

Stage-based evaluation pipeline for generative molecular design: filters, retrosynthesis checks, docking, pose validation, reports, CLI/TUI.