Which numbering scheme should I use?

IMGT when you want one scheme for every chain type, including T-cell receptors; Kabat or Chothia when a paper, patent or modelling tool expects them. All six give the same residues different positions, so pick the one your downstream work reads.

Why do the CDR borders differ between definitions?

IMGT, Kabat, Chothia and North draw the loops by different criteria (sequence variability, structure, clustering), so the same residue can be CDR under one and framework under another. The panel shows where every definition puts each border.

Does it handle scFvs and Fab sequences?

Yes. Each variable domain in the input is found, typed and numbered on its own.

Is anything uploaded?

No. Numbering, germline search and the liability scan run in the browser.

Analyze

Antibody Numbering

Number an antibody the way the literature does. Paste or select a protein, and every antibody or T-cell-receptor variable domain in it is found, typed by chain and species, and numbered under IMGT, Kabat, Chothia, Martin, AHo or Wolfguy. The frameworks and CDRs come out under the definition you choose, the closest V and J germlines with their identity, and a scan for the chemical liabilities that sink candidates in development, each placed at its IMGT position. Checked against ANARCI on every chain of its own benchmark.

Open Antibody Numbering Open the Antibody numbering sample

Trastuzumab Fab and its HER2 target from PDB 1N8Z, for IMGT/Kabat/Chothia numbering, germlines and sequence liabilities

What it does

01Finds every antibody and T-cell-receptor variable domain in a protein, including both domains of an scFv
02Chain type (heavy, kappa, lambda, TCR alpha, beta, gamma, delta) and species from profile HMMs
03Numbering under IMGT, Kabat, Chothia, Martin, AHo or Wolfguy, with insertion codes
04Framework and CDR regions under the IMGT, Kabat, Chothia or North CDR definition, and CDR3 length
05Closest V and J germline genes with percent identity, in the species you allow
06Sequence liabilities with their IMGT positions and regions: deamidation, isomerization, N-glycosylation sequons, Asp-Pro cleavage, Met and Trp oxidation, unpaired or missing cysteines
07Humanization toward the closest human germline, and CDR grafting onto a human framework
08A domain map drawn residue by residue, and CDR-coloured 3D structure ribbons

What it runs on

A faithful port of ANARCI, checked against the real tool on every distinct chain of its own PDB benchmark (3,747 numberable chains), with CDR definitions checked against abnumber; each liability rule cites its published source. The panel runs on @ports/antibody, @ports/protein.

Everything runs client-side. Files are parsed in the browser and projects are saved on this machine, so an unpublished sequence is never uploaded.

Questions

Which numbering scheme should I use?: IMGT when you want one scheme for every chain type, including T-cell receptors; Kabat or Chothia when a paper, patent or modelling tool expects them. All six give the same residues different positions, so pick the one your downstream work reads.
Why do the CDR borders differ between definitions?: IMGT, Kabat, Chothia and North draw the loops by different criteria (sequence variability, structure, clustering), so the same residue can be CDR under one and framework under another. The panel shows where every definition puts each border.
Does it handle scFvs and Fab sequences?: Yes. Each variable domain in the input is found, typed and numbered on its own.
Is anything uploaded?: No. Numbering, germline search and the liability scan run in the browser.

Also searched as

antibody numbering online · IMGT numbering · Kabat numbering · Chothia numbering · find CDRs in an antibody sequence · CDR annotation · ANARCI alternative · antibody germline assignment · antibody liability scan · deamidation and isomerization motifs · scFv domain numbering · TCR variable domain numbering · antibody humanization · CDR grafting