---
title: Literature Passport
description: >-
  Research intelligence report combining entity extraction, relationship
  mapping, trend analysis, and competitor activity.
tags:
  - passports
  - literature
---

The Literature Passport answers "what's the research landscape for this topic?" by mining published research into extracted entities, mapped relationships, temporal trends, and competitor activity. You give it a single search query, and generate it when you want the assembled intelligence for a domain rather than raw search hits. Running BioBERT entity extraction and reading the papers yourself yields the same signal, but the passport pulls from that fixed source set and layers relationship mapping, trend analysis, and novelty scoring on top. See [what a passport is](/docs/passports-explained) for the model behind these.

## Literature Passport

The Literature Passport mines published research to extract entities, map relationships, identify emerging trends, and track competitor activity in a research domain.

> The Literature Passport requires the **Grow plan or higher**.

## Input

- **Format**: Search query (research topic, 2–200 characters), optional project ID
- **Duration**: 10–30 seconds
- **Access**: Grow plan or higher

## Data Sources Aggregated

### BioBERT Entities

Named entities extracted with confidence scores:
- Targets and proteins
- Genes and mutations
- Diseases and conditions
- Drugs and compounds
- Mechanisms and pathways
- Biomarkers

Each entity includes confidence score and document source references.

### Entity Relationships

Entity-to-entity connections identified from text:
- `associated_with_disease` — target–disease links
- `in_pathway` — pathway membership
- `targeted_by` — drug–target associations
- `interacts_with` — protein–protein interactions
- `homolog` — evolutionary relationships

### Temporal Trends

Publication recency analysis with 6-month cutoff:
- **Increasing** — >60% of mentions in last 6 months
- **Stable** — balanced distribution
- **Decreasing** — majority of mentions older than 6 months

### Company Activity

Competitor intelligence from entity text pattern matching:
- Company mentions and frequency
- Target focus areas per company
- Recent activity signals

## Novelty Scoring

Entities receive a novelty score:
- **40% frequency component** — inverse of mention count (rarer = more novel)
- **60% trend component** — recent mention ratio (increasing trend = more novel)

## Sections Displayed

1. **Entity Landscape** — top entities per type with mention counts and trends
2. **Relationships** — entity network visualization
3. **Emerging Targets** — novel targets with increasing momentum
4. **Competitor Activity** — pharma company research focus areas
5. **Research Gaps** — underexplored areas identified by low mention count + high relevance

## Canvas Visualization

- **Card size**: 300×250px
- **Border color**: Brown (#795548)
- **Tier 2 metrics**: Entity count by type, relation count, recent publication count

## Chat Usage

### Example Prompts

- "Build a literature passport for CRISPR gene editing"
- "What's the research landscape for GLP-1 receptor agonists?"
- "Generate a literature report on EGFR C797S resistance mechanisms"

### Workflow Chaining

- **Accepts from**: PubMed search results, free-text research queries
- **Feeds into**: Target Passport (extracted targets), Compound Passport (extracted compounds)