Home
AbMart
Member Benefits
Tools
Resource
About
Group Sites
Language
tool-banner
Tool Catalog
Home>Tool Catalog>

AbHuGrafter

AbHuGrafter
AbHuGrafter
Lead Antibody Optimization
2026-04-10
Try Now

1 Introduction

AbHuGrafter is an antibody humanization tool based on CDR Grafting technology, supporting humanization of nanobodies and conventional antibodies. By transplanting the CDR regions of antibodies to the framework regions of human templates and then performing back-mutations at specific sites, it achieves antibody humanization. The tool integrates multiple scoring metrics, including AbSeek humanization assessment, T20 scoring, structural similarity analysis, and more.

Main Features

Grafting Humanization

  • Dual Antibody Type Support: Supports humanization of nanobodies and conventional antibodies
  • Specified Template Grafting: Such as using h-NbBcII10_FGLA1 template (nanobody only), with continuous addition of new templates
  • Automatic Template Selection: Automatically selects the most similar human template
  • Vernier Site Retention: Preserves Kabat Vernier sites to maintain structural stability
  • Hallmark Site Back-mutation: Supports back-mutation of specified key sites
  • preADA Site Processing: Automatically fixes preADA sites to V amino acid
  • V/J Gene Specification: Supports separate specification of V and J genes for VH and VL chains
  • Extra Disulfide Bond Detection: Automatically detects and processes extra_disulfide sites

Multi-dimensional Scoring System

  • AbSeek Humanization Assessment: Humanization scoring based on AbSeek server
  • T20 Scoring: Thermostability evaluation
  • CDR Side Chain RMSD Analysis: Side chain structural similarity of CDR regions
  • Backbone RMSD Analysis: Protein backbone structural similarity

Structure Prediction and Comparison

  • Based on AbSeek Server: Performs structure prediction and similarity analysis
    • Nanobodies: Algorithm will call AbSeek-ImmuneBuilder (AntiBody) tool for prediction
    • Conventional antibodies: Algorithm will call AbSeek-ImmuneBuilder (NanoBody) tool for prediction
  • PDB Structure File Generation: Outputs complete PDB structure files
  • Structure Alignment Analysis: Supports backbone and side chain alignment

2 Parameters

Parameter Description
Protein type Antibody type (antibody or nanobody)
Heavy Chain Sequence Heavy chain (VH) variable region amino acid sequence
Light Chain Sequence Light chain (VL) variable region amino acid sequence, required only when type is antibody; hidden when type is nanobody
Antibody VH Template VH chain humanization template type (such as Human Germline, h-NbBcII10_FGLA, etc.), where h-NbBcII10_FGLA only supports nanobodies
Antibody VL Template VL chain humanization template type (Human Germline), available only when antibody-type is antibody
VH V Gene VH chain V gene (such as IGHV1-18), can only be used when Antibody VH Template is "Human Germline". When set to None, algorithm automatically finds the most matching gene
VH J Gene VH chain J gene (such as IGHJ1), can only be used when Antibody VH Template is "Human Germline". When set to None, algorithm automatically finds the most matching gene
VL V Gene VL chain V gene (such as IGKV1-39), can only be used when Antibody VL Template is "Human Germline". When set to None, algorithm automatically finds the most matching gene
VL J Gene VL chain J gene (such as IGKJ1), can only be used when Antibody VL Template is "Human Germline". When set to None, algorithm automatically finds the most matching gene
Backmutations Back-mutation site types (vernier, hallmark, preADA, extra_disulfide)
Scoring Metrics Scoring metrics list, multiple selection, all selected by default

Supplementary Description of Back-mutation Types

  • vernier sites:
    These are the following sites in Kabat numbering2.
    • VH: [2, 27, 28, 29, 30, 47, 48, 49, 67, 69, 71, 73, 78, 93, 94, 103]
    • VK and VL: [2, 4, 35, 36, 46, 47, 48, 49, 64, 66, 68, 69, 71, 98]
  • hallmark sites:
    • IMGT numbering2 positions [42, 49, 50, 52]
  • preADA sites:
    • IMGT numbering2 position 12, back-mutated to V amino acid
  • extra_disulfide sites:
    • IMGT numbering2 positions [55, 108]. For some nanobodies with longer CDR3, the IMGT positions 55 and 1082 are cysteine residues, forming additional disulfide bonds that play an important role in maintaining their structure.

3 Reference

[1] Cécile Vincke, Remy Loris, Dirk Saerens, Sergio Martinez-Rodriguez, Serge Muyldermans, Katja Conrath, General Strategy to Humanize a Camelid Single-domain Antibody and Identification of a Universal Humanized Nanobody Scaffold*,Journal of Biological Chemistry, Volume 284, Issue 5, 2009, Pages 3273-3284, ISSN 0021-9258, https://doi.org/10.1074/jbc.M806889200.

[2] James Dunbar, Charlotte M. Deane, ANARCI: antigen receptor numbering and receptor classification, Bioinformatics, Volume 32, Issue 2, January 2016, Pages 298–300, https://doi.org/10.1093/bioinformatics/btv552