Beyond Binding – Fast, Scalable koff Predictions Powered by SILCS and AI
SilcsBio hosted a webinar on January 15th, 2026 showcasing how SILCS‑Kinetics is transforming the study of ligand binding–unbinding kinetics by enabling rapid, scalable predictions of ligand dissociation rates (koff) using a combination of physics‑based modeling and machine learning.
About SILCS‑Kinetics
Understanding ligand dissociation behavior is essential for predicting drug residence time, optimizing efficacy, and designing ligands with favorable kinetic profiles. Traditional approaches for estimating koff often rely on long-timescale molecular dynamics simulations, limiting throughput and practicality.
SILCS‑Kinetics overcomes these challenges by integrating SILCS free-energy information with machine learning to:
- Identify and characterize ligand dissociation pathways.
- Generate free‑energy profiles that describe unbinding energetics.
- Predict ligand koff values at scale, without long MD simulations.
- Provide atomic and functional‑group insights into dissociation mechanisms.
If you missed the live webinar, don’t worry!
The full recording is available to watch here:
Key Takeaways
Kinetics‑Driven Design:
How SILCS‑Kinetics enables prediction of koff and residence time, central parameters for drug efficacy.
High‑Throughput koff Predictions:
A scalable workflow combining SILCS free‑energy profiles with ML models to efficiently evaluate large ligand sets.
Pathway Enumeration & Energetics:
How SILCS identifies likely dissociation routes and quantifies their contributions.
Case Studies and Validation:
Performance across 329 ligands and 13 protein targets, demonstrating robustness and broad applicability.
Integration with SILCS Workflows:
How SILCS‑Kinetics complements SILCS‑FragMaps, SILCS‑Monte Carlo docking, and other components of the platform.
