Antibody modeling

OpenDDE integrates ImmuneBuilder to predict antibody 3D structures directly from heavy and light chain amino acid sequences. The resulting structures include annotated CDR loops (H1–H3, L1–L3) for analysis.

How it works

1. Navigate to the Antibody page from the sidebar
2. Paste your heavy chain and light chain amino acid sequences
3. Click Predict structure
4. The ImmuneBuilder service generates a 3D PDB structure
5. View the result in the interactive 3D viewer with CDR loops color-coded

CDR loop annotation

Complementarity-Determining Regions (CDRs) are the most variable parts of the antibody and are responsible for antigen binding. OpenDDE highlights all six CDR loops:

• H1, H2, H3 — Heavy chain CDRs
• L1, L2, L3 — Light chain CDRs

H3 is typically the most variable and important for antigen specificity.

About ImmuneBuilder

ImmuneBuilder is developed by the Oxford Protein Informatics Group (OPIG). It uses deep learning to predict antibody structures with accuracy comparable to AlphaFold, but specialized for immunoglobulin folds.

Use cases

• Analyze therapeutic antibody candidates
• Study CDR loop conformations for engineering
• Compare predicted structures with experimental data
• Visualize paratope-epitope interactions

← Complex prediction · AI assistant →