textToKnowledgeGraph

textToKnowledgeGraph utilizes Large Language Models to extract molecular interactions from scientific articles and convert them into Biological Expression Language (BEL) format for structured knowledge graph generation.

Key Features:

• BEL Extraction: Extracts molecular interactions from scientific articles directly in Biological Expression Language (BEL) format.
• Large Language Model Integration: Leverages LLMs trained on diverse literature to enable accurate extraction of contextual information across sentences.
• Command-line Interface: Provides a command-line interface for BEL extraction from scientific articles.
• Cytoscape Web Integration: Includes an interactive application within Cytoscape Web for extraction and exploration of knowledge graphs.
• Fine-tuning Dataset: Provides a dataset of extractions that have been both computationally and manually reviewed to support future fine-tuning efforts.

Scientific Applications:

• Knowledge Graph Construction: Structures and analyzes biological information by representing data and improving queries across heterogeneous datasets.
• Molecular Interaction Extraction: Extracts complex molecular relationships from unstructured scientific literature that require awareness of indirectly described concepts.
• Biological Network Exploration: Enables exploration of molecular interaction networks within Cytoscape Web.

Methodology:

textToKnowledgeGraph uses Large Language Models to extract interactions from individual publications directly in Biological Expression Language (BEL).