Designing transferable transition state guided collective variable via interpretable machine learning model for enhanced sampling: A case study on polymer collapse transition

Key Points

• This research aims to develop an interpretable machine learning model that optimizes collective variables for enhanced sampling of polymer collapse transitions.
• Utilization of enhanced sampling methods to study biophysical processes.
• Implementation of machine learning techniques to design collective variables.
• Computational simulations to observe polymer collapse transitions.
• Identification of effective collective variables improves the efficiency of sampling.
• Achieved higher accuracy in modeling polymer behaviors during collapse transitions.