SSP-Stat

Secondary Structure Prediction using GOR and SVM

SSP-Stat is a Python-based toolkit designed for protein secondary structure prediction. It implements both the GOR (Garnier–Osguthorpe–Robson) statistical method and Support Vector Machine (SVM) classifiers to predict secondary structures from amino acid sequences.

---

Features

• GOR Implementation: Utilizes information theory-based statistical methods for secondary structure prediction.
• SVM Classifiers: Employs machine learning techniques for enhanced prediction accuracy.
• Dataset Preparation: Tools to process and prepare datasets for training and evaluation.
• Performance Evaluation: Includes metrics such as Q3 accuracy and Segment Overlap (SOV) scores.

---

Installation

1. Clone the Repository:

   git clone https://github.com/vincenzo-palmacci/SSP-Stat.git
   cd SSP-Stat

2. Set Up a Virtual Environment (Optional but Recommended):

   python3 -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install Dependencies:

   pip install -r requirements.txt

---

Usage

1. Prepare the Dataset

Use the make-dataset.py script to process raw protein sequence data into a format suitable for training and evaluation.

python make-dataset.py --input raw_data.fasta --output processed_dataset.pkl

2. Train the GOR Model

Navigate to the gor-tools directory and execute the training script:

cd gor-tools
python train_gor.py --dataset ../processed_dataset.pkl --model gor_model.pkl

3. Train the SVM Model

Navigate to the svm-tools directory and execute the training script:

cd ../svm-tools
python train_svm.py --dataset ../processed_dataset.pkl --model svm_model.pkl

4. Evaluate Model Performance

Use the performance.py script to assess the accuracy of your models:

python performance.py --model gor_model.pkl --test_data test_dataset.pkl

This will output metrics such as Q3 accuracy and SOV scores.

---

Project Structure

SSP-Stat/
├── gor-tools/           # GOR model implementation
├── svm-tools/           # SVM model implementation
├── stats/               # Statistical analysis tools
├── Dataset.py           # Dataset handling class
├── make-dataset.py      # Script for dataset preparation
├── performance.py       # Performance evaluation script
├── sov.py               # SOV score calculation
├── requirements.txt     # Python dependencies
└── README.md            # Project documentation

---

Requirements

• Python 3.6 or higher
• Required Python packages listed in requirements.txt