Target Passport

The Target Passport answers "is this protein worth pursuing?" by assembling its metadata, structures, druggability screening, and literature context into one target viability report with a completeness score. You give it a single UniProt ID, and generate it when you're evaluating a target rather than inspecting one source in isolation. Querying UniProt, PDB, your docking results, and BioBERT literature yourself surfaces the same facts, but the passport pulls from that fixed set of sources and presents them keyed to the one protein. See what a passport is for the model behind these.

Target Passport

The Target Passport aggregates protein metadata, structural data, druggability assessment, and literature context into a comprehensive target viability report.

Input

• Format: UniProt ID (e.g., P04637 for p53)
• Duration: 5–20 seconds
• Access: All users

Data Sources Aggregated

Protein Metadata

Name, sequence length, organism, and functional annotations from UniProt.

Structure Data

Experimental PDB structures and AlphaFold predictions with quality scores. Includes resolution data for experimental structures and pLDDT confidence for predicted structures.

Druggability Assessment

Docking results filtered by this target:

• Compounds screened: Total number tested
• Hit rate: Percentage with affinity < -7.0 kcal/mol
• Best affinity: Most favorable binding energy
• Median affinity: Typical binding energy across screened compounds

Literature Context

BioBERT-extracted entities from published literature:

• Associated diseases and pathways
• Known drugs targeting this protein
• Related targets in the same family
• Total mention count across publications

Scoring

• Completeness score: Weighted 0.25 per section
• Structure: 0.5 if sequence present + 0.5 if experimental structure available
• Druggability: Scales with compounds screened (saturates at 100)
• Literature: Scales with mention count (saturates at 50)

Sections Displayed

1. Structure — sequence info + 3D structure availability (experimental and predicted)
2. Druggability — screening statistics, hit rate, affinity distribution
3. Literature Context — disease associations, pathways, known drugs, related targets
4. Completeness Overview — data coverage across all sections

Canvas Visualization

• Card size: 300×250px
• Border color: Brown (#795548)
• Tier 2 metrics: Gene name, druggability score, PDB structure count, literature reference count

Chat Usage

Example Prompts

• "Generate a target passport for P04637" (p53)
• "Create a target profile for EGFR (P00533)"
• "How druggable is CDK4? Run a target passport"

Workflow Chaining

• Accepts from: UniProt search, BLAST (hit proteins), BioBERT (extracted targets)
• Feeds into: Structure prediction (ESMFold), docking (DiffDock), compound passport for known binders