Tutorials

These tutorials show how to use the OpenEye Structural Biology Floes, understand their results, and connect to other OpenEye floes downstream. The key tasks are listed below:

  1. Prepare a system for starting a WEMD simulation using cryo-EM data.

  2. Refine an initial structure from ab initio structure builders to a consensus or target 3D cryo-EM map to obtain a structural ensemble for cryptic pocket prediction.

  3. Sample conformational space around the initial structure using the mean maps and eigenmaps from 3D variability analysis (3DVA) or RECOVAR analysis of cryo-EM data to obtain a structural ensemble for cryptic pocket prediction.

  4. Extract the best structure ensembles for a series of cryo-EM maps after a WEMD simulation.

  5. Select a final state and obtain the transition path from the initial structure.

  6. Run OpenEye Cryptic Pocket Detection Floes to predict allosteric pocket sites.

Note

This package uses cryo-EM data to guide the WEMD simulation. The accuracy of the simulation depends on several factors:

  • Quality of force fields: These include proteins and other molecules in the simulation system. Generally, force fields for DNA and RNA biomolecules and metal ions are less accurate than those of proteins.

  • Quality of cryo-EM maps: To resolve a side-chain conformation of a residue, a cryo-EM map might need a high resolution of less than 3 Å. But changes of secondary structures or domain motions can be observed on medium or low resolution maps, which can still be used to guide WEMD simulations for large conformational transitions.

  • Convergence of simulation: WEMD can greatly accelerate conformation sampling compared to traditional MD simulation. For a large and complicated system, however, a well-converged simulation might not be enough. Theoretically, infinite simulation time may be necessary to reach a global equilibrium. So all simulations of large systems are conditional to the initial structures and the simulation time. But a local equilibrium around important minima can be reached in a finite time by examining the evolution of progress coordinates or the Kullback-Leibler divergence of the associated probability density distributions. For those large, complicated systems, the guidance of cryo-EM becomes more significant in helping limit the conformational space to explore.