Automated design of collective variables using supervised machine learning
2018/09/07 by Mohammad M. Sultan, Vijay S. Pande · 145 citations
Computer Science · #Advanced Multi-Objective Optimization Algorithms #Artificial intelligence #Computer science #Evolutionary Algorithms and Applications #Machine learning #Software Engineering Research
paper · doi:10.1063/1.5029972
published in The Journal of Chemical Physics 149(9), 094106 (American Institute of Physics)
openalex publication_date 2018/09/07 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/01
Abstract
Selection of appropriate collective variables (CVs) for enhancing sampling of molecular simulations remains an unsolved problem in computational modeling. In particular, picking initial CVs is particularly challenging in higher dimensions. Which atomic coordinates or transforms there of from a list of thousands should one pick for enhanced sampling runs? How does a modeler even begin to pick starting coordinates for investigation? This remains true even in the case of simple two state systems and only increases in difficulty for multi-state systems. In this work, we solve the “initial” CV problem using a data-driven approach inspired by the field of supervised machine learning (SML). In particular, we show how the decision functions in SML algorithms can be used as initial CVs (SMLcv) for accelerated sampling. Using solvated alanine dipeptide and Chignolin mini-protein as our test cases, we illustrate how the distance to the support vector machines’ decision hyperplane, the output probability estimates from logistic regression, the outputs from shallow or deep neural network classifiers, and other classifiers may be used to reversibly sample slow structural transitions. We discuss the utility of other SML algorithms that might be useful for identifying CVs for accelerating molecular simulations.
Citations
Cited by
- Coarse Graining Molecular Dynamics with Graph Neural Networks
- Deep learning of committor and explainable artificial intelligence analysis for identifying reaction coordinates
- Improving the Exploration of High Dimensional Free Energy Landscape by a Combination of Temperature Accelerated Sliced Sampling and Parallel Biasing
- Revealing the Atomistic Mechanism of Rare Events in Molecular Dynamics
- Entangled Schrödinger Bridge Matching
- Enhanced Sampling in the Age of Machine Learning: Algorithms and Applications
- A Bayesian approach for extracting free energy profiles from cryo-electron microscopy experiments using a path collective variable
- Computing rates and understanding unbinding mechanism in host-guest systems
- Learning Structure, Energy, and Dynamics: A Survey of Artificial Intelligence for Protein Dynamics
- Action-Minimization Meets Generative Modeling: Efficient Transition Path Sampling with the Onsager-Machlup Functional
- Anncolvar: Approximation of Complex Collective Variables by Artificial Neural Networks for Analysis and Biasing of Molecular Simulations. [europepmc]
- Time-Lagged t-Distributed Stochastic Neighbor Embedding (t-SNE) of Molecular Simulation Trajectories. [europepmc]
- Data-Driven Molecular Dynamics: A Multifaceted Challenge. [europepmc]
- Coarse graining molecular dynamics with graph neural networks. [europepmc]
- TorchMD: A Deep Learning Framework for Molecular Simulations. [europepmc]
- A survey of multiscale modeling: Foundations, historical milestones, current status, and future prospects. [europepmc]
- How to Determine Accurate Conformational Ensembles by Metadynamics Metainference: A Chignolin Study Case. [europepmc]
- A Bayesian approach to extracting free-energy profiles from cryo-electron microscopy experiments. [europepmc]
- New perspectives in cancer drug development: computational advances with an eye to design. [europepmc]
- Deep learning the slow modes for rare events sampling. [europepmc]
- Enhanced sampling without borders: on global biasing functions and how to reweight them. [europepmc]
- Pandemic drugs at pandemic speed: infrastructure for accelerating COVID-19 drug discovery with hybrid machine learning- and physics-based simulations on high-performance computers. [europepmc]
- Pepsin-like aspartic proteases (PAPs) as model systems for combining biomolecular simulation with biophysical experiments. [europepmc]
- Protein Function Analysis through Machine Learning. [europepmc]
- Collective variable discovery in the age of machine learning: reality, hype and everything in between. [europepmc]
- Temporally Coherent Backmapping of Molecular Trajectories From Coarse-Grained to Atomistic Resolution. [europepmc]
- Collective Variables for Crystallization Simulations-from Early Developments to Recent Advances. [europepmc]
- Constructing Collective Variables Using Invariant Learned Representations. [europepmc]
- Reaction Coordinates for Conformational Transitions Using Linear Discriminant Analysis on Positions. [europepmc]
- OneOPES, a Combined Enhanced Sampling Method to Rule Them All. [europepmc]
- Perspectives on Ligand/Protein Binding Kinetics Simulations: Force Fields, Machine Learning, Sampling, and User-Friendliness. [europepmc]
- Combining stochastic resetting with Metadynamics to speed-up molecular dynamics simulations. [europepmc]
- Advanced computational approaches to understand protein aggregation. [europepmc]
- Correlating enzymatic reactivity for different substrates using transferable data-driven collective variables. [europepmc]
- Enhanced Sampling in the Age of Machine Learning: Algorithms and Applications. [europepmc]
Related