Add dataset card and link to paper

README.md

@@ -1,3 +1,61 @@
 ---
 license: apache-2.0
+task_categories:
+- graph-ml
+tags:
+- biology
+- protein-protein-interaction
+- multimodal
 ---
+
+This repository contains the processed datasets and pre-trained checkpoints for the paper [Enhancing Protein-Protein Interaction Prediction with Hierarchical Motif-based Multimodal Protein Embedding](https://huggingface.co/papers/2606.02629).
+
+**Code:** [https://github.com/yzf-code/MMM-PPI](https://github.com/yzf-code/MMM-PPI)
+
+### Dataset Overview
+
+MMM-PPI is a Hierarchical Motif-based Multi-Modal protein Encoder for PPI Prediction. It addresses limitations in existing PPI prediction by modeling the hierarchical organization of proteins and integrating sequence, structure, and function modalities across three scales: micro-scale (residues), meso-scale (spatially-informed motifs), and macro-scale (protein level).
+
+### Download Instructions
+
+You can download the processed datasets and pre-trained checkpoints using the following methods:
+
+**Using huggingface_hub**
+```bash
+pip install huggingface_hub
+huggingface-cli download yzf1102/MMM-PPI --repo-type dataset --local-dir ./hf_assets
+```
+
+**Using git lfs**
+```bash
+git lfs install
+git clone https://huggingface.co/datasets/yzf1102/MMM-PPI hf_assets
+```
+
+### Repository Structure
+
+After downloading, organize the files as follows to match the project structure:
+
+```
+MMM-PPI/
+├── ckpt/                          # Pre-trained model checkpoints
+├── data/                          # Dataset files
+├── motif_detection/               # Motif detection scripts and pre-computed results
+├── Classifier_model.py            # PPI classifier model
+├── dataloader.py                  # Data loading and preprocessing
+├── environment.yml                # Conda environment specification
+├── main.py                        # Main entry point for training/evaluation
+├── pretrain_Pair_wise_Encoder.py  # Pair-wise encoder pre-training 
+├── run.sh                         # Example shell script for running experiments
+└── utils.py                       # Utility functions
+```
+
+> 💡 Make sure the `ckpt/` and `data/` folders are placed at the **root** of the repository.
+
+### Usage
+
+The simplest way to reproduce the experiments is via the provided shell script:
+
+```bash
+bash run.sh
+```