Rewrite tutorials from user conversation perspective

Major shift in documentation approach:
- Rewrite tutorials to show natural language conversations with Claude Desktop
- Replace code-centric examples with realistic user dialogues
- Focus on what users actually say vs technical commands
- Show ChatSpatial responses and automatic analysis
- Emphasize no-coding-required approach
- Include realistic conversation flows and follow-up questions

Key changes:
- getting_started.md: Add conversation examples showing ease of use
- basic_spatial_analysis.md: Complete rewrite as natural dialogue
- Shift from 'write this code' to 'ask this question'
- Show actual ChatSpatial responses and insights
- Emphasize biological interpretation over technical details

This addresses the core issue that users interact through conversation,
not code, making tutorials much more relevant to actual usage.

Author: cafferychen777

docs/getting_started.md

@@ -1,77 +1,144 @@
 # Getting Started with ChatSpatial
 
-This comprehensive guide will help you install, configure, and run your first spatial transcriptomics analysis with ChatSpatial.
+Transform your spatial transcriptomics analysis from complex coding to natural conversation! This guide shows you how to set up ChatSpatial and start analyzing your data through simple questions in Claude Desktop.
 
-## Prerequisites
+## 🎯 What You'll Achieve
 
-Before installing ChatSpatial, ensure you have:
+By the end of this guide, you'll be able to:
+- 💬 **Ask questions** about your spatial data in plain English
+- 🧬 **Analyze tissue architecture** without writing code
+- 🎨 **Generate beautiful visualizations** automatically
+- 🔬 **Discover biological insights** through conversation
 
-- **Python 3.10+** (recommended: 3.10 or 3.11)
-- **Conda** or **Miniconda** for environment management
-- **Git** for cloning the repository
-- **8GB+ RAM** for typical spatial datasets
-- **Claude Desktop** or another MCP-compatible client
+## 🚀 Quick Start (5 Minutes)
 
-### 📥 Download Claude Desktop
+### Step 1: Get Claude Desktop
 
-If you don't have Claude Desktop yet:
+**New to Claude?** No problem!
 
-1. **Visit**: [claude.ai](https://claude.ai)
-2. **Download**: Claude Desktop for your operating system
-3. **Sign up**: Create an Anthropic account if needed
+1. 🌐 **Visit**: [claude.ai](https://claude.ai)
+2. 📱 **Download**: Claude Desktop for your computer
+3. 👤 **Sign up**: Create your free Anthropic account
 
-### 🎥 Learn About MCP First
+### Step 2: Install ChatSpatial
 
-New to Model Context Protocol? Watch this quick introduction:
+**Don't worry - this is easier than it looks!**
+
+Open your terminal/command prompt and run these commands:
+
+```bash
+# Create a new environment (like a clean workspace)
+conda create -n chatspatial python=3.11
+conda activate chatspatial
+
+# Get ChatSpatial
+git clone https://github.com/cafferychen777/ChatSpatial.git
+cd ChatSpatial
+pip install -e .
+```
+
+### Step 3: Connect to Claude Desktop
+
+**This is where the magic happens!**
+
+1. **Find your Python path**:
+   ```bash
+   conda activate chatspatial
+   which python
+   ```
+   Copy this path (something like `/Users/yourname/miniconda3/envs/chatspatial/bin/python`)
+
+2. **Configure Claude Desktop**:
+   - **Mac**: Open `~/Library/Application Support/Claude/claude_desktop_config.json`
+   - **Windows**: Open `%APPDATA%\Claude\claude_desktop_config.json`
+   - **Linux**: Open `~/.config/Claude/claude_desktop_config.json`
+
+3. **Add this configuration** (replace the Python path with yours):
+   ```json
+   {
+     "mcpServers": {
+       "chatspatial": {
+         "command": "/your/python/path/here",
+         "args": ["-m", "chatspatial"],
+         "env": {}
+       }
+     }
+   }
+   ```
+
+4. **Restart Claude Desktop**
+
+### 🎉 You're Ready!
+
+Look for the 🔬 ChatSpatial tools in Claude Desktop. You should see spatial analysis tools available!
+
+## 🎥 New to MCP? Watch This First!
 
 [![MCP Introduction](https://img.shields.io/badge/▶️_Watch-What_is_MCP?-blue?style=flat-square&logo=youtube)](https://www.youtube.com/watch?v=sfCBCyNyw7U)
 
-**Additional Resources:**
-- 📚 [Official MCP Documentation](https://modelcontextprotocol.io)
+**Learn more:**
+- 📚 [What is MCP?](what_is_mcp.md) - Detailed explanation
 - 📰 [Anthropic's MCP Announcement](https://www.anthropic.com/news/model-context-protocol)
 - 💻 [MCP GitHub Repository](https://github.com/modelcontextprotocol)
 
-## Installation
+## 💬 Your First Conversation
 
-### Step 1: Clone the Repository
+Once ChatSpatial is connected, here's how easy it is to analyze your data:
 
-```bash
-git clone https://github.com/cafferychen777/ChatSpatial.git
-cd ChatSpatial
-```
+---
 
-### Step 2: Create Python Environment
+**👤 You:** "Hi! I'm new to spatial transcriptomics. Can you help me analyze my Visium data?"
 
-```bash
-# Create a new conda environment
-conda create -n chatspatial python=3.10
-conda activate chatspatial
-```
+**🤖 Claude:** "I'd love to help you explore your spatial transcriptomics data! ChatSpatial makes it easy - no coding required. Just tell me what you'd like to discover about your tissue.
 
-### Step 3: Install Dependencies
+What kind of sample do you have, and where is your data file located?"
 
-```bash
-# Install ChatSpatial in development mode
-pip install -e .
+**👤 You:** "I have a mouse brain sample. The file is at `/Users/myname/data/brain_sample.h5ad`"
 
-# Verify installation
-chatspatial --help
-```
+**🤖 Claude:** "Perfect! Let me load your mouse brain data and take a look at what we're working with."
 
-### Step 4: Install Optional Dependencies
+*[ChatSpatial loads and analyzes the data]*
 
-For full functionality, install additional packages:
+**🤖 Claude:** "Excellent! Your mouse brain sample looks great:
+- 2,845 spots across the tissue
+- 28,691 genes measured
+- High-quality spatial coordinates included
 
-```bash
-# R dependencies for scType
-conda install -c conda-forge r-base r-essentials
+What would you like to explore? I can help you:
+- 🧠 Identify different brain regions
+- 🔬 Find cell types in each area
+- 📊 Discover spatially variable genes
+- 🎨 Create beautiful visualizations
+- 💬 Analyze cell-cell communication
 
-# Additional spatial analysis tools
-pip install SpaGCN STAGATE liana cellphonedb
+What interests you most?"
 
-# GPU support (optional, for faster processing)
-pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118
-```
+**👤 You:** "I'd love to see the different brain regions!"
+
+**🤖 Claude:** "Great choice! Let me identify the spatial domains in your brain tissue..."
+
+*[Analysis happens automatically]*
+
+**🤖 Claude:** "Amazing! I found 6 distinct brain regions:
+- Cortical layers (outer brain)
+- Hippocampus (memory center)
+- White matter (connecting fibers)
+- Ventricular zone (stem cells)
+- Corpus callosum (brain bridge)
+- Boundary regions
+
+Here's a beautiful spatial map showing each region in different colors! 🧠✨"
+
+---
+
+### 🎯 See How Natural It Is?
+
+- ❌ **No complex commands** to remember
+- ❌ **No parameter tuning** required
+- ❌ **No coding experience** needed
+- ✅ **Just ask questions** in plain English!
+- ✅ **Get instant insights** about your tissue
+- ✅ **Beautiful visualizations** automatically generated
 
 ## MCP Configuration