Add model documentation

README.md

@@ -0,0 +1,106 @@
+---
+license: mit
+tags:
+- drug-discovery
+- protein-ligand-binding
+- binding-kinetics
+- deep-learning
+- computational-biology
+- bioinformatics
+library_name: pytorch
+datasets:
+- kineticX
+---
+
+# BiCoA-Net: Bidirectional Co-Attention Network
+
+## Model Description
+
+BiCoA-Net predicts protein-ligand dissociation rate constants (k_off) using bidirectional co-attention mechanisms between protein and ligand representations.
+
+**Key Features:**
+- Predicts binding kinetics (k_off) for drug-target interactions
+- Uses ESM-2 protein embeddings + MolFormer ligand embeddings
+- Bidirectional co-attention fusion mechanism
+- Trained on curated KineticX datasets
+
+## Quick Start
+```python
+from huggingface_hub import hf_hub_download
+import torch
+
+# Download model weights
+model_path = hf_hub_download(
+    repo_id="Daisyli95/BiCoA-Net",
+    filename="pytorch_model.pt"
+)
+
+# Load model (FP16 format)
+state_dict = torch.load(model_path, map_location='cpu')
+
+# Convert to FP32 for inference (recommended)
+state_dict_fp32 = {k: v.float() if v.dtype == torch.float16 else v 
+                   for k, v in state_dict.items()}
+
+# Load into your model architecture
+model.load_state_dict(state_dict_fp32)
+model.eval()
+```
+
+## Model Details
+
+- **Architecture**: ESM-2 (650M) + MolFormer + Bidirectional Co-Attention
+- **Training Data**: PDBbind v2020 + Custom kinetics data
+- **Format**: PyTorch FP16 (1.80 GB)
+- **Parameters**: ~960M
+- **Input**: 
+  - Protein sequence (FASTA)
+  - Ligand SMILES string
+- **Output**: Predicted log(k_off) value
+
+## Performance
+
+- Concordance Index (C-index): [Add your metrics]
+- Pearson Correlation: [Add your metrics]
+- Test on held-out GPCR targets: [Add your metrics]
+
+## Usage Example
+```python
+# Assuming you have BiCoA-Net model class defined
+from your_model import BiCoANet
+
+# Initialize model
+model = BiCoANet()
+
+# Load pretrained weights
+state_dict = torch.load(model_path, map_location='cpu')
+state_dict = {k: v.float() for k, v in state_dict.items()}
+model.load_state_dict(state_dict)
+model.eval()
+
+# Predict
+protein_seq = "MSLQKEVQKL..."
+ligand_smiles = "CC(C)Cc1ccc(cc1)C(C)C(O)=O"
+
+with torch.no_grad():
+    prediction = model(protein_seq, ligand_smiles)
+    print(f"Predicted log(k_off): {prediction.item()}")
+```
+
+## Training Details
+
+- Optimizer: AdamW
+- Learning Rate: 1e-4
+- Batch Size: 32
+- Epochs: 100
+- Loss Function: MSE on log(k_off)
+
+
+
+## License
+
+MIT License - Free for academic and commercial use.
+
+## Contact
+
+For questions or issues, please open an issue on the GitHub repository or contact the authors.