TML-MP

TML-MP predicts changes in protein folding free energy caused by single point mutations using a topology-based representation to assess mutation-induced stability changes.

Key Features:

• Topology-Based Framework: Reduces the geometric complexity and degrees of freedom of protein structures by representing them through topological descriptors.
• Element-Specific Persistent Homology: Applies element-specific persistent homology to capture topological features while preserving essential biological information.

Scientific Applications:

• Globular Proteins: Predicts stability changes in globular proteins with a Pearson correlation coefficient of 0.82 and an RMSE of 0.92 kcal/mol on a test set of 350 mutation samples.
• Membrane Proteins: Predicts stability changes in membrane proteins with a Pearson correlation coefficient of 0.57 and an RMSE of 1.09 kcal/mol in 5-fold cross-validation on 223 mutation samples, and outperforms current empirical methods by 84%.

Methodology:

Uses a topology-based analysis to reduce structural complexity and element-specific persistent homology to retain critical biological information.