update user guide

implement top level organization
seperate topics in different md files

docs/user_guide.md

@@ -1,63 +1,30 @@
 # AxoWise User Guide
 
 ## Overview
-The **User Guide** provides a step-by-step walkthrough of how to use AxoWise to analyze biological networks, detect communities, and extract functional insights using AI-supported techniques.
-
-## Table of Contents
-1. [Input Table: Listing Proteins](#input-table-listing-proteins)
-2. [Detecting Protein Communities with Protein Network](#detecting-protein-communities-with-protein-network)
-3. [Functional Exploration with Function Network](#functional-exploration-with-function-network)
-4. [Extracting Relevant Knowledge from Publications](#extracting-relevant-knowledge-from-publications)
-5. [AI-Supported Knowledge Extraction](#ai-supported-knowledge-extraction)
-
-## 1. Input Table: Listing Proteins
-To begin an analysis, users need to provide an **input table** that lists proteins of interest.
-- Format: The table should be in CSV or TSV format.
-- Required columns:
-  - `Protein_ID`: Unique identifier for the protein.
-  - `Protein_Name`: Common or scientific name.
-  - `Source`: Data source (e.g., UniProt, KEGG).
-- Example Input Table:
-
-  | Protein_ID | Protein_Name  | Source  |
-  |------------|--------------|---------|
-  | P12345     | ExampleProt1 | UniProt |
-  | Q67890     | ExampleProt2 | KEGG    |
-
-## 2. Detecting Protein Communities with Protein Network
-AxoWise enables users to identify **protein communities** based on network connections.
-- The **protein network** is constructed using interaction data.
-- Algorithms used for community detection include:
-  - Louvain clustering
-  - Label Propagation
-  - Modularity-based approaches
-- The result shows clusters of related proteins that may be functionally associated.
-
-## 3. Functional Exploration with Function Network
-After detecting protein communities, AxoWise allows for **functional exploration** through a function network:
-- Functional terms (e.g., Gene Ontology terms) are linked to proteins.
-- The function network visualizes relationships between biological functions.
-- Helps in understanding functional modules within detected protein communities.
-
-## 4. Extracting Relevant Knowledge from Publications
-AxoWise integrates literature mining to extract relevant knowledge.
-- Uses **matching abstracts** from biological publications.
-- Retrieves abstracts linked to proteins or functional terms.
-- Helps users explore relevant scientific literature automatically.
-
-## 5. AI-Supported Knowledge Extraction
-AxoWise leverages **AI models** to enhance knowledge extraction:
-- **Named Entity Recognition (NER)** for identifying key biological terms.
-- **Semantic Similarity Analysis** to find related studies.
-- **Topic Modeling** to categorize literature into functional themes.
-- AI-powered summarization to highlight key insights from scientific texts.
-
-## Summary
-This guide provides a structured approach to leveraging AxoWise for:
-✔ Protein community detection
-✔ Functional network exploration
-✔ Literature mining
-✔ AI-supported insights
-
-For further details, refer to individual sections or visit the [AxoWise GitHub Repository](https://github.com/BackofenLab/AxoWise).
+AxoWise is designed to help you develop contextual biological insights using protein interaction data, functional relationships, and literature-based knowledge extraction. It leverages multiple interconnected networks and advanced AI-driven techniques to provide meaningful insights.
+
+### Networks in AxoWise
+AxoWise integrates three key networks:
+- **Protein Network**: Identifies and clusters protein communities based on interaction data, helping uncover functional modules.
+- **Function Network**: Maps functional relationships between proteins to explore biological roles and interactions.
+- **Publication Network**: Extracts relevant abstracts and knowledge from literature, linking biological concepts with scientific publications.
+
+### Graph Retrieval Augmented Generation (GRAG)
+AxoWise employs **Graph Retrieval Augmented Generation (GRAG)** to enhance biological insight extraction:
+- **Graph Retrieval**: Retrieves structured biological knowledge from the integrated networks, ensuring context-aware data exploration.
+- **Augmented Generation**: Uses AI-powered summarization and text generation to provide actionable insights, integrating findings from the protein, function, and publication networks into a cohesive narrative.
+AxoWise is designed to help you develop contextual biological insights using protein interaction data, functional relationships, and literature-based knowledge extraction.
+
+## Quick Start
+To get started with AxoWise quickly, follow the [Quick Start Guide](quick_start.md).
+
+## Detailed Topics
+1. **[Input Data](input_data.md)** → Learn about supported input formats.
+2. **[Protein Network](protein_network.md)** → Detecting and clustering protein communities.
+3. **[Function Network](function_network.md)** → Exploring functional relationships between proteins.
+4. **[Publication Network](publication_network.md)** → Extracting knowledge from literature.
+5. **[AI Extraction](ai_extraction.md)** → AI-supported extraction of actionable insights.
+6. **[Troubleshooting](troubleshooting.md)** → Common issues and solutions.
+
+For a full explanation of how AxoWise works, refer to the sections above.