alan-ai/abmelt-thermostability

🧬 AbMelt: Antibody Thermostability Predictor

Predict antibody thermostability properties using molecular dynamics descriptors and machine learning.

About

AbMelt combines multi-temperature molecular dynamics simulations with machine learning to predict key antibody thermostability properties:

• —Aggregation Temperature (Tagg): Measures tendency to aggregate
• —Melting Temperature (Tm): Thermal unfolding temperature
• —Melt Onset Temperature (Tmon): Initial unfolding temperature

Features

• —🔬 Physics-based: Uses molecular dynamics simulation descriptors
• —🎯 Accurate: Outperforms sequence-only methods
• —🚀 Fast: Instant predictions from MD descriptors
• —📊 Interpretable: Clear thermostability assessments

Usage

1. 1.Input your molecular dynamics descriptors:
2. 2.RMSF of CDR regions at 400K
3. 3.Radius of gyration standard deviation at 400K
4. 4.All-temperature lambda descriptor

1. 1.Get instant predictions for all three thermostability properties

1. 1.Interpret results using the provided guidance

Reference

Rollins, Z.A., et al. "AbMelt: Learning antibody thermostability from molecular dynamics." Biophysical Journal (2024).

GitHub Repository