Antibody & Biologics · Deane Lab / Oxford
Paragraph
Predicts antibody binding interface residues from structure to support humanisation and immunogenicity studies.
What Paragraph is
Paragraph is an antibody modelling tool developed by the Deane Lab at the University of Oxford. It identifies which residues form the binding interface directly from protein structural data alone, without requiring experimental complex structures.
In antibody engineering workflows, Paragraph is used within Corollary to steer humanisation edits away from essential binding sites and to flag potentially immunogenic surface patches across the molecule.
Key facts
- Version
- 1.0
- Domain
- Antibody
- Tasks
- Epitope prediction, paratope prediction
- Input format
- PDB (antibody structure)
- Output format
- Residue-level probability table
- GPU memory
- 8 GB
What you can use it for
- Paratope prediction
- Calculates per-residue binding probabilities to locate paratope residues directly from an uncomplexed antibody structure.
- Guided antibody humanisation
- Identifies binding interface residues so sequence modifications can be directed away from the paratope to maintain binding affinity.
- Surface immunogenicity profiling
- Flags surface patches on the antibody structure that may present immunogenic liabilities.
Run it here
No installation and no GPU of your own: fill in the inputs and it runs on our servers. Results stay in your browser unless you sign in.
Not runnable yet
This model isn't available to run right now.Run this model
Paragraph
Strengths and limits
Good at
- Predicts binding interface residues using solely the antibody structure.
- Provides granular, per-residue paratope probability outputs in tabular form.
Watch out for
- Requires a coordinate structure in PDB format rather than accepting raw amino acid sequences.
- Demands 8 GB of GPU memory for execution.
Inputs and outputs
You provide
Antibody structure (PDB)
You get back
Per-residue paratope probability.
Licence
Commercial use permitted · BSD-3-Clause
The BSD-3-Clause licence allows use in commercial research and products.
Questions
Can I use Paragraph commercially?
Commercial use permitted. The BSD-3-Clause licence allows use in commercial research and products.
How do I run Paragraph without installing anything?
Use the form under “Run it here” on this page: fill in the inputs and press Run model. It runs on our servers on 1× A100 16GB, typically in 3.0s, and a run costs about 14 credits. Visitors without an account get 1,000 credits a month.
What input data does Paragraph require?
It accepts antibody coordinate structures in PDB format.
What are the primary outputs of the model?
It generates a table displaying paratope probabilities calculated for each individual residue.
How is Paragraph applied in antibody engineering?
It is used within workflows such as Corollary to ensure humanising mutations do not disturb paratope residues and to detect immunogenic surface patches.
Written from Paragraph's own documentation. Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.