Skip to main content
skill CX Codex

tooluniverse-gpcr-structural-pharmacology — Claude Code skill

The tooluniverse-gpcr-structural-pharmacology skill researches GPCR pharmacology, receptor structures, ligand binding, and antibody interfaces. Use it for drug discovery and mutation questions.

Apache-2.0 241 installs ★ 1,798 updated 2026-10-24

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add mims-harvard/ToolUniverse --skill tooluniverse-gpcr-structural-pharmacology -a claude-code
Open on GitHub

What tooluniverse-gpcr-structural-pharmacology does

The skill uses GPCRdb to identify a receptor, retrieve receptor details, list known ligands, classify their pharmacology, and inspect binding affinities. It can also collect available structures and mutation data linked to receptor function or ligand binding.

Structural work combines PDB records with PDBePISA for interface pairs, predicted biological assemblies, and per-chain surface analysis. Antibody-related questions use SAbDab to find structures, CDR annotations, and chain details. A valid GPCRdb entry name is required before receptor queries, so the workflow starts with GPCRdb_list_proteins rather than guessing a slug.

Pharmacology results distinguish agonists, antagonists, inverse agonists, partial agonists, and biased agonists. The skill also relates active or inactive receptor states to the structures used for ligand interpretation. It is available for Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, and OpenCode.

What it helps with

  • find a GPCR by family or receptor class
  • classify known ligands and compare their affinities
  • map receptor mutations to structural data
  • inspect protein interfaces in a PDB entry
  • retrieve antibody structures for an antigen

Example prompts for the agent

  • Find the GPCRdb entry for the beta-2 adrenergic receptor and list its ligands.
  • Compare active and inactive structures for adrb2_human, then explain which records fit an agonist-binding analysis.
  • Search SAbDab for antibody structures against the target antigen and analyze PDB 6W41.

How to install tooluniverse-gpcr-structural-pharmacology

1Skills CLI (Vercel)

npx skills add mims-harvard/ToolUniverse --skill tooluniverse-gpcr-structural-pharmacology -a claude-code

2Manually: put the files in place

~/.claude/skills/tooluniverse-gpcr-structural-pharmacology/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/tooluniverse-gpcr-structural-pharmacology/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about tooluniverse-gpcr-structural-pharmacology

How do I install tooluniverse-gpcr-structural-pharmacology in Claude Code?

Run npx skills add mims-harvard/ToolUniverse --skill tooluniverse-gpcr-structural-pharmacology -a claude-code, then check ~/.claude/skills/tooluniverse-gpcr-structural-pharmacology/SKILL.md.

What does tooluniverse-gpcr-structural-pharmacology do?

It researches GPCRs, ligands, mutations, and structures through GPCRdb, while SAbDab handles antibody structure questions. PDBePISA supplies interface and assembly analysis.

Why must the receptor be found in GPCRdb first?

GPCRdb tools expect an entry name such as adrb2_human, rather than a gene symbol or UniProt accession. Use GPCRdb_list_proteins first and pass its result to later queries.

Source text

SKILL.md · Apache-2.0
---
name: tooluniverse-gpcr-structural-pharmacology
description: GPCR receptor pharmacology — agonist/antagonist/inverse-agonist/biased-agonist classification, GPCRdb structural data, receptor-ligand binding analysis, antibody-target interface (SAbDab). Use for GPCR drug discovery, biased-agonism analysis, receptor subtype selectivity questions, and orthosteric vs allosteric pocket characterization.
disable-model-invocation: true
---

# GPCR and Structural Pharmacology Research

**GPCR pharmacology**: agonist vs antagonist vs inverse agonist vs biased agonist — each has different clinical implications. Biased agonism (preferential G-protein vs β-arrestin signaling) can separate efficacy from side effects; for example, G-protein-biased opioid agonists aim to retain analgesia while reducing β-arrestin-mediated respiratory depression. Always classify retrieved ligands by their pharmacological type, not just their chemical structure. Receptor state (active vs inactive crystal structure) determines which ligands and mutations are interpretable — an inactive-state structure is appropriate for antagonist binding analysis, active-state for agonist-bound complexes. Generic GPCR numbering (Ballesteros-Weinstein) enables cross-receptor mutation comparison; always report positions in this system alongside sequence positions.

**LOOK UP DON'T GUESS**: never assume GPCRdb entry names (e.g., `adrb2_human`) or PDB IDs — always use `GPCRdb_list_proteins` to find the correct entry name and `GPCRdb_get_structures` to confirm available structures.

Research skill integrating GPCRdb (GPCR receptor biology), SAbDab (antibody structures), and PDBePISA (protein interface analysis) to support structural pharmacology, antibody engineering, and GPCR-targeted drug discovery.

**KEY PRINCIPLES**:
1. **Receptor-first** — Identify GPCR entry name before any GPCRdb queries
2. **Ligand classification** — Distinguish agonists, antagonists, partial agonists, biased agonists
3. **Structure-guided** — Pair GPCRdb mutation data with PDB structures via PDBePISA
4. **Antibody context** — Use SAbDab for therapeutic antibody structure retrieval and CDR analysis
5. **English-first queries** — Use standard receptor names (e.g., "beta-2 adrenergic receptor") in searches; convert to GPCRdb entry names for API calls

---

## When to Use

Apply when user asks:
- "What ligands are known for [GPCR receptor]?"
- "What crystal structures exist for [receptor]?"
- "Find antibody structures targeting [antigen]"
- "Analyze the protein-protein interface in PDB [ID]"
- "What mutations affect [GPCR] function or pharmacology?"
- "Which GPCRs are in the [family] family?"
- "What are the CDR loops in antibody PDB [ID]?"
- "What is the biological assembly for [PDB ID]?"

---

## Tool Parameter Reference (CRITICAL)

| Tool | Key Parameters | Notes |
|------|---------------|-------|
| `GPCRdb_get_protein` | `protein` | GPCRdb entry name (e.g., `adrb2_human`), NOT gene symbol or UniProt accession |
| `GPCRdb_list_proteins` | `family` (optional), `protein_class` (optional) | Lists all GPCRs; filter by family slug (e.g., `"adrenoceptors"`) OR by human-readable class name via `protein_class` (e.g., `"chemokine receptors"`, `"opioid receptors"`) |
| `GPCRdb_get_structures` | `protein` (optional), `state` (optional) | `state`: `"active"`, `"inactive"`, `"intermediate"` |
| `GPCRdb_get_ligands` | `protein` | Returns agonists, antagonists, biased ligands with affinities |
| `GPCRdb_get_mutations` | `protein` | Returns mutation effects on receptor function and ligand binding |
| `SAbDab_search_structures` | `query` | Antigen name, species, or keywords; returns browse URL + metadata |
| `SAbDab_get_structure` | `pdb_id` | 4-character PDB code (e.g., `"6W41"`); returns CDR annotations |
| `SAbDab_get_summary` | (no required params) | Database statistics and summary |
| `PDBePISA_get_interfaces` | `pdb_id` | 4-character PDB code; returns all interface pairs with buried area |
| `PDBePISA_get_assemblies` | `pdb_id` | Predicted biological assemblies from crystal packing |
| `PDBePISA_get_monomer_analysis` | `pdb_id` | Per-chain solvent-accessible surface area (SASA) breakdown |

### GPCRdb Entry Name Format

GPCRdb uses its own entry name format: `{receptor_slug}_{species}`. Common examples:
- Beta-2 adrenergic receptor: `adrb2_human`
- Beta-1 adrenergic receptor: `adrb1_human`
- Mu-opioid receptor: `oprm1_human`
- Dopamine D2 receptor: `drd2_human`
- Glucagon-like peptide-1 receptor: `glp1r_human`
- CXCR4 chemokine receptor: `cxcr4_human`

If entry name is unknown, use `GPCRdb_list_proteins()` to browse and find the correct slug. You can also filter by receptor class using the `protein_class` parameter with a human-readable name — e.g., `GPCRdb_list_proteins(protein_class="chemokine receptors")` — instead of the numeric family slug. Both `family` and `protein_class` are accepted and serve overlapping purposes; prefer `protein_class` when the user provides a receptor class name.

---

## Workflow Overview

```
Phase 1: Receptor Identification (for GPCR queries)
  -> GPCRdb_list_proteins: find receptor family and entry name
  -> GPCRdb_get_protein: receptor details, family, species

Phase 2: Ligand Landscape
  -> GPCRdb_get_ligands: all known ligands by pharmacology class
  -> Cross-reference with ChEMBL/PubChem for chemical properties

Phase 3: Structural Data
  -> GPCRdb_get_structures: available PDB/EMDB structures with resolution
  -> PDBePISA_get_interfaces: interface analysis on best structure
  -> PDBePISA_get_assemblies: biological assembly determination

Phase 4: Mutation & Pharmacology Data
  -> GPCRdb_get_mutations: pharmacological mutation map
  -> Compare to ligand binding sites from structure

Phase 5: Antibody Structures (for antibody queries)
  -> SAbDab_search_structures: find structures by antigen
  -> SAbDab_get_structure: CDR annotations, chain details
  -> PDBePISA_get_interfaces: antibody-antigen interface analysis
```

---

## Phase 1: GPCR Receptor Identification

```python
# List all GPCRs in a family to find entry name (by slug)
family_list = GPCRdb_list_proteins(family="adrenoceptors")

# Filter by human-readable class name (new -- preferred when user says e.g. "chemokine receptors")
chemokine_list = GPCRdb_list_proteins(protein_class="chemokine receptors")

# Browse all GPCRs (no family filter)
all_gpcrs = GPCRdb_list_proteins()

# Get detailed protein info once you have the entry name
receptor = GPCRdb_get_protein(protein="adrb2_human")
# Returns: family classification, endogenous ligands, tissue expression,
#          GPCRdb-specific annotations, sequence features
```

## Phase 2: Ligand Landscape

```python
# Get all known ligands for a GPCR
ligands = GPCRdb_get_ligands(protein="adrb2_human")
# Returns: ligand names, types (agonist/antagonist/partial/biased/allosteric),
#          binding affinities (Ki, IC50, EC50), references

# Ligand type classification:
# - Agonist: activates receptor
# - Antagonist/Inverse agonist: blocks or suppresses receptor
# - Partial agonist: submaximal activation
# - Biased agonist: selective signaling (Gs vs. beta-arrestin bias)
# - Positive/Negative allosteric modulator (PAM/NAM)
```

After retrieving ligands from GPCRdb, optionally cross-reference with:
- `PubChem_get_CID_by_compound_name(compound_name=ligand_name)` — get CID, SMILES
- `ChEMBL_search_molecules(query=ligand_name)` — get ChEMBL ID, bioactivity data

## Phase 3: Structural Data

```python
# Get available crystal/cryo-EM structures
structures = GPCRdb_get_structures(protein="adrb2_human", state="inactive")
# state options: "active", "inactive", "intermediate" (omit for all)
# Returns: PDB IDs, resolution, ligand in structure, publication info

# Analyze a specific structure's interfaces
interfaces = PDBePISA_get_interfaces(pdb_id="2rh1")  # adrb2 inactive structure
# Returns: interface pairs, buried solvent-accessible area (BSA),
#          interface residues, hydrogen bonds, salt bridges

# Determine biological assembly
assemblies = PDBePISA_get_assemblies(pdb_id="2rh1")
# Returns: predicted oligomeric state, assembly stability score,
#          subunit composition

# Per-chain SASA breakdown
monomers = PDBePISA_get_monomer_analysis(pdb_id="2rh1")
# Returns: accessible/buried surface area per chain
```

**Interface Analysis Interpretation**:
- BSA > 1500 Ų: Strong interface (likely biologically relevant)
- BSA 800-1500 Ų: Moderate interface
- BSA < 800 Ų: Weak or crystal contact

## Phase 4: Mutation Data

```python
# Get all mutations characterized for a GPCR
mutations = GPCRdb_get_mutations(protein="adrb2_human")
# Returns: mutation positions (generic GPCR numbering), effects on:
#   - Expression/folding
#   - Ligand binding (affinity changes)
#   - G-protein coupling
#   - Receptor activation

# Generic GPCR numbering (Ballesteros-Weinstein):
# e.g., 3.32 = position 32 in TM helix 3 — conserved across GPCR classes
```

## Phase 5: Antibody Structure Retrieval

```python
# Search SAbDab for antibody structures by antigen
results = SAbDab_search_structures(query="EGFR", limit=20)
# Returns: browse URL + metadata table of matching structures

# Get detailed annotations for a specific antibody structure
structure = SAbDab_get_structure(pdb_id="1IQD")
# Returns: VH/VL chain IDs, CDR1-3 (Kabat/IMGT), antigen info,
#          heavy/light chain types, resolution

# Get database overview
summary = SAbDab_get_summary()
# Returns: total structures, species breakdown, antigen coverage stats
```

**CDR Analysis**:
- CDR-H3 is most variable and typically dominates antigen contact
- CDR length distribution: SAbDab provides Kabat, Chothia, and IMGT numbering
- After retrieving SAbDab structure, use `PDBePISA_get_interfaces(pdb_id=...)` to compute antibody-antigen buried surface area

---

## Common Research Patterns

### Pattern 1: GPCR Drug Target Profiling
```
Input: GPCR name (e.g., "GLP-1 receptor")
Flow: GPCRdb_list_proteins -> find "glp1r_human" ->
      GPCRdb_get_protein (receptor details) ->
      GPCRdb_get_ligands (approved + investigational drugs) ->
      GPCRdb_get_structures (available PDB structures) ->
      PDBePISA_get_interfaces on best structure ->
      GPCRdb_get_mutations (pharmacological mutants)
Output: Complete GPCR pharmacology profile with structural context
```

### Pattern 2: Antibody-Antigen Interface Analysis
```
Input: Target antigen (e.g., "PD-L1") or specific PDB code
Flow: SAbDab_search_structures(query="PD-L1") ->
      SAbDab_get_structure(pdb_id="best hit") (CDR annotations) ->
      PDBePISA_get_interfaces(pdb_id=...) (buried area, key contacts) ->
      PDBePISA_get_assemblies (assembly context)
Output: CDR sequences, epitope contact residues, interface energetics
```

### Pattern 3: GPCR Family Survey
```
Input: Drug class question (e.g., "beta-adrenergic receptors")
Flow: GPCRdb_list_proteins(family="adrenoceptors") ->
      GPCRdb_get_protein per receptor (adrb1/2/3) ->
      GPCRdb_get_ligands per receptor (selectivity landscape) ->
      GPCRdb_get_structures per receptor (structural coverage)
Output: Family-wide selectivity map, structural availability, ligand classes
```

### Pattern 4: Structure Interface Characterization
```
Input: PDB code
Flow: PDBePISA_get_assemblies (oligomeric state) ->
      PDBePISA_get_interfaces (all interface pairs ranked by BSA) ->
      PDBePISA_get_monomer_analysis (per-chain surface burial)
Output: Biologically relevant assembly, key interface residues, buried areas
```

---

## Tool Combinations with Other Skills

This skill complements other ToolUniverse skills:

| Goal | This skill provides | Complement with |
|------|--------------------|--------------------|
| GPCR drug discovery | Receptor/ligand/structure data | `tooluniverse-binder-discovery` for virtual screening |
| Antibody engineering | SAbDab structure + CDR data | `tooluniverse-antibody-engineering` for optimization |
| Variant impact on GPCR | GPCRdb mutation effects | `tooluniverse-variant-functional-annotation` for ACMG |
| Target validation | GPCR expression, ligand data | `tooluniverse-drug-target-validation` |
| PDB structure analysis | PDBePISA interfaces | `tooluniverse-protein-structure-retrieval` for RCSB/PDBe |

---

## Fallback Chains

| Primary Tool | Fallback | Use When |
|-------------|----------|----------|
| `GPCRdb_get_protein` | UniProt search + PubMed | Entry name unknown or non-GPCR target |
| `GPCRdb_get_ligands` | ChEMBL bioactivity search | Receptor not in GPCRdb |
| `GPCRdb_get_structures` | RCSB PDB text search | Structures not yet in GPCRdb |
| `SAbDab_search_structures` | RCSB PDB antibody search | Antigen not indexed in SAbDab |
| `PDBePISA_get_interfaces` | PDBe graph API | PDBePISA returns no interfaces |

---

## Completeness Checklist

For GPCR profiling:
- [ ] Entry name resolved via `GPCRdb_list_proteins` or `GPCRdb_get_protein`
- [ ] Receptor family and class documented
- [ ] Ligand landscape retrieved with pharmacology types
- [ ] Available structures listed with resolution and state
- [ ] Best structure analyzed with PDBePISA (interfaces + assembly)
- [ ] Mutation data retrieved for pharmacological context

For antibody structure:
- [ ] SAbDab search run with antigen name
- [ ] Best structure retrieved with `SAbDab_get_structure`
- [ ] CDR1-3 sequences extracted for VH and VL chains
- [ ] Antibody-antigen interface analyzed with PDBePISA
- [ ] Buried surface area and key contact residues documented

---

## Key References

- GPCRdb: https://gpcrdb.org — standardized GPCR data with generic numbering
- SAbDab: https://opig.stats.ox.ac.uk/webapps/newsabdab/sabdab — structural antibody database
- PDBePISA: https://www.ebi.ac.uk/pdbe/pisa — protein interface analysis

© mims-harvard · Apache-2.0 · mims-harvard/ToolUniverse

Similar skills

All skills for Science and research →

From the same set136

The whole set →
tooluniverse-protein-structure-prediction tooluniverse-protein-structure-prediction predicts protein structures from amino acid sequences and gathers evidence to assess the models. Use it to examine a new protein or explore how a variant may affect its structure. CX Codex +1 245installs tooluniverse-chemical-sourcing tooluniverse-chemical-sourcing finds commercial chemical suppliers across ZINC, Enamine, eMolecules, and Mcule after validating structures via PubChem or ChEMBL. CX Codex +1 243installs tooluniverse-clinical-data-integration tooluniverse-clinical-data-integration builds a drug-safety review from FDA labels, FAERS, pharmacogenomics, trials, and literature. Use it for pharmacovigilance and regulatory assessment. CX Codex +1 242installs tooluniverse-proteomics-data-retrieval tooluniverse-proteomics-data-retrieval — a skill that finds public proteomics datasets across MassIVE, ProteomeXchange, PRIDE Archive, and PDC. Search by species, keyword, accession, or protein, and get back metadata or download links. CX Codex +1 241installs tooluniverse-sequence-analysis tooluniverse-sequence-analysis — a skill for fetching sequences from NCBI, Ensembl, and UniProt, plus running FASTQ quality control and coverage calculations. CX Codex +1 241installs tooluniverse-epigenomics-chromatin The tooluniverse-epigenomics-chromatin skill examines histone marks, chromatin accessibility, and TF binding with data from ENCODE and related epigenomics resources. CX Codex +1 241installs tooluniverse-plant-genomics tooluniverse-plant-genomics queries plant gene structures, metabolic pathways, and orthologs using PlantReactome, KEGG, and Ensembl Plants. Useful for analyzing polyploid crop genomes and plant stress responses. CX Codex +1 240installs tooluniverse-structural-proteomics tooluniverse-structural-proteomics is a Claude Code skill that gathers structural and proteomics data to assess drug targets. It supports binding-site, ligand, and model-confidence analysis. CX Codex +1 240installs

Project sponsors

Companies that keep this analytics open