Update README with deployment instructions

- HF push workflow
- Git credentials setup
- Space settings reference
- Local dev instructions
- File structure overview

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

README.md

@@ -1,43 +1,84 @@
 ---
 title: CRISPR Array Detection
 emoji: 🧬
-colorFrom: blue
-colorTo: green
+colorFrom: gray
+colorTo: gray
 sdk: docker
 pinned: false
 license: mit
 ---
 
-# CRISPR Array Detection
+# crispr-detect
 
-Detect CRISPR arrays in DNA sequences using a BERT-based deep learning model.
+BERT-based CRISPR array detection in prokaryotic genomes.
 
 ## Model
 
-- **Architecture**: 24-layer BERT transformer
-- **Parameters**: ~430 million
-- **Input**: DNA sequence (minimum 1000 bp)
-- **Output**: Per-position CRISPR probability (0-1)
+| | |
+|---|---|
+| architecture | BERT, 24 layers, 768 hidden, 430M params |
+| input | DNA sequence (min 1000 bp) |
+| output | per-position probability (0-1) |
 
-## Features
+## Deployment
 
-- **Predict**: Get CRISPR probability scores for each nucleotide position
-- **Detect**: Automatically find CRISPR array regions above a threshold
-- **Embeddings**: Extract hidden state vectors for visualization
+### Push changes
 
-## Usage
+```bash
+cd /vol/hpcprojects/pmuench/crispr_tool/crispr-hf-space
+git add -A
+git commit -m "description"
+git push
+```
 
-1. Paste your DNA sequence (FASTA format or raw, minimum 1000 bp)
-2. Click "Predict" to get per-position scores
-3. Or use "Detect Regions" for automated CRISPR calling
+### Git credentials (first time)
 
-## Citation
+```bash
+git config --global credential.helper store
+huggingface-cli login
+# paste token from https://huggingface.co/settings/tokens
+```
 
-Based on [CRISPRArrayDetection](https://github.com/Ziyu-Mu/CRISPRArrayDetection)
+### Clone fresh
+
+```bash
+git clone https://huggingface.co/spaces/genomenet/crispr-array-detection
+```
+
+### Space settings (HuggingFace web UI)
+
+- SDK: Docker
+- Hardware: T4 GPU
+- Visibility: Public
+
+### Model weights
+
+Hosted at: https://huggingface.co/pmuench3/crispr-bert-model
+
+Downloaded automatically via `huggingface_hub` at startup.
+
+## Local dev
+
+```bash
+pip install -r requirements.txt
+python app.py
+# http://localhost:7860
+```
+
+## Files
+
+```
+├── app.py              # gradio app
+├── inference/
+│   ├── model_loader.py # model download
+│   ├── tokenizer.py    # sequence validation
+│   └── inference.py    # prediction
+├── Dockerfile
+└── requirements.txt
+```
 
 ## Acknowledgements
 
-- Ziyu Mu, Master's Thesis, HZI BIFO
-- DFG SPP 2141 (Geschäftszeichen MC 172)
+- Ziyu Mu (HZI BIFO)
+- DFG SPP 2141 (MC 172)
 - BMBF GenomeNet
-- Helmholtz Centre for Infection Research