---
title: Compound Passport
description: >-
  Aggregated compound profile combining ADMET, synthesis, vendor, docking, and
  molecular property data.
tags:
  - passports
  - compound
---

The Compound Passport answers "what do I need to know about this molecule?" in one view — its physicochemical properties, ADMET predictions, synthesis routes, vendor availability, and docking behavior, assembled and scored for completeness. You give it 1–10 SMILES strings, and generate it when you want the whole profile of a compound rather than one facet at a time. Running ADMET-AI, AiZynthFinder, ZINC, DiffDock, and RDKit yourself gives you the same underlying data, but you'd stitch and weight the results by hand; the passport orchestrates the collection and presents them together. See [what a passport is](/docs/passports-explained) for the model behind these.

## Compound Passport

The Compound Passport aggregates data from five sources into a single comprehensive profile for one or more small molecules.

## Input

- **Format**: 1–10 SMILES strings
- **Duration**: 5–30 seconds
- **Access**: All users

## Data Sources Aggregated

### ADMET Panel

~40 prediction endpoints from ADMET-AI, categorized by absorption, distribution, metabolism, excretion, and safety. Each prediction includes a confidence score and severity flag (green/yellow/red).

### Synthesis Routes

Retrosynthetic analysis from AiZynthFinder with route scoring, predicted yield, reagent details, and step count. Routes link to vendor availability for building blocks.

### Vendor Availability

Commercial sourcing with pricing, stock quantity, lead time, and supplier information, where vendor data exists for the compound. Shows which vendors have the compound in stock.

### Docking Results

Binding poses from DiffDock with affinity statistics per target protein. Includes confidence scores and pose count for each docked complex.

### Molecular Properties

RDKit-computed physicochemical properties:

| Property | Description |
|----------|-------------|
| MW | Molecular weight |
| LogP | Partition coefficient (lipophilicity) |
| TPSA | Topological polar surface area |
| HBD | Hydrogen bond donors |
| HBA | Hydrogen bond acceptors |
| Rotatable bonds | Molecular flexibility |
| QED | Quantitative estimate of drug-likeness |

## Scoring

- **Completeness score**: 0–100% based on data coverage across all 5 sections
- **Section weights**: 0.2 each (equal weighting)
- **ADMET score**: Average confidence across prediction categories

## Sections Displayed

1. **Molecular Properties** — physicochemical descriptors and drug-likeness
2. **ADMET** — absorption, distribution, metabolism, excretion, and toxicity with color-coded flags
3. **Synthesis Routes** — retrosynthetic routes with vendor sourcing links
4. **Vendor Availability** — pricing table with stock status and lead times
5. **Docking Summary** — binding affinity statistics per target

## Canvas Visualization

- **Card size**: 300×250px
- **Border color**: Brown (#795548)
- **Tier 2 metrics**: ADMET flag count, vendor count, synthesis route count, docked target count
- **Provenance**: Data lineage tracking shows which jobs contributed data

## Chat Usage

### Example Prompts

- "Generate a compound passport for CC(=O)Oc1ccccc1C(=O)O" (aspirin)
- "Run compound passports on my top 5 hits from the ChEMBL search"
- "Create a full profile for gefitinib and erlotinib — compare ADMET"

### Workflow Chaining

- **Accepts from**: PubChem, ZINC, ChEMBL, REINVENT, GenMol (any tool that outputs SMILES)
- **Feeds into**: Export/share as PDF or comparison table