skills.rest
mims-harvard
mims-harvardOFFICIAL
Explore all skills by this author →
Website𝕏@marinkazitnikGitHub

tooluniverse-antibody-engineering

Guides therapeutic antibody humanization, affinity maturation, and developability assessment using ToolUniverse databases.

1.7k|254|Updated Mar 3, 2025
Search Tags:#developability#humanization#antibody engineering#immunogenicity#biologics#affinity maturation#cdr grafting
One-click install
npx skills add https://github.com/mims-harvard/ToolUniverse --skill tooluniverse-antibody-engineering
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: tooluniverse-antibody-engineering
Source: https://github.com/mims-harvard/ToolUniverse/tree/main/plugins/tooluniverse/skills/tooluniverse-antibody-engineering
Command: npx skills add https://github.com/mims-harvard/ToolUniverse --skill tooluniverse-antibody-engineering

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Optimizing a therapeutic antibody from a preclinical lead to a clinical candidate requires coordinating humanization, affinity maturation, developability scoring, immunogenicity prediction, and manufacturing assessment across many databases and tools. This Skill orchestrates that entire workflow with evidence-graded scoring and structured reports.

Core Features & Use Cases

  • Sequence Humanization: Aligns VH/VL sequences to IMGT human germlines, designs CDR grafting with Vernier zone backmutations, and scores framework humanness.
  • Developability & Immunogenicity Assessment: Computes aggregation-prone regions, pI, hydrophobic patches, and PTM liability sites, plus IEDB-based T-cell epitope risk scoring.
  • Clinical Precedent Benchmarking: Queries TheraSAbDab and SAbDab for approved antibodies and structural precedents against the same target.
  • Use Case: Given a mouse anti-PD-L1 antibody sequence, produce a ranked set of humanized variants with developability scores, immunogenicity risk, and an experimental validation plan in a structured markdown report.

Quick Start

Ask your AI agent to humanize your antibody VH and VL sequences against a target antigen using the ToolUniverse antibody engineering workflow and generate the optimization report.

Frequently Asked Questions about tooluniverse-antibody-engineering

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I humanize a mouse antibody sequence?▼

Provide the VH and VL sequences and target antigen, then the workflow identifies closest human germlines via IMGT, grafts CDRs onto selected frameworks, and designs Vernier zone backmutations. It outputs at least two humanized variants with humanness scores and developability metrics.

What tools does this antibody engineering workflow use?▼

It uses ToolUniverse tools: IMGT for germline genes, TheraSAbDab for clinical precedents, SAbDab for antibody structures, IEDB for epitope prediction, and UniProt for target information. A bundled developability script computes aggregation, pI, and hydrophobic patches from sequence.

Why do IMGT, SAbDab, and TheraSAbDab tool calls fail with a parameter error?▼

These are SOAP-based tools that require an operation parameter as the first argument, such as operation="search_genes" for IMGT_search_genes. Omitting it causes a validation error stating operation is a required property.

Can this skill predict antibody thermal stability and expression titer?▼

No, Tm and expression titer are not computable from sequence alone and require external ML predictors. The skill only reports these values if such a tool is actually run, and explicitly warns against fabricating them.

What should I do when TheraSAbDab returns no results for my target?▼

Try alternative target names, since TheraSAbDab requires exact name matching. For example, query PD-L1 as PDL1, CD274, or B7-H1, and HER2 as ERBB2, until clinical precedents are found.

Related Skills

Explore More Education & Research Skills→
PharMolixPharMolix
1.1k

antibody-structure-prediction-tfold

Predict antibody and nanobody structures from sequences using tfold-based models.

Community
Advanced
biomaps-infrabiomaps-infra

service-adaptyv

Submit protein sequences for experimental validation via a cloud laboratory API.

Official
Advanced
PKU-YuanGroupPKU-YuanGroup
288

chai1

Predict protein, nucleic-acid, and small-molecule complex structures with Chai-1.

Official
Intermediate
yjcyxkyyjcyxky

boltzgen

Design protein binders with diffusion-based generative models.

Community
Advanced