Monte Carlo sampling with linear inverse kinematics for simulation of protein flexible regions

A Monte Carlo linear inverse-kinematics method for the simulation of protein chains with fixed ends is introduced. It includes backbone bond-angle bending and simultaneous loop and ring closure to allow full proline ring flexibility.
An obstacle to linear null-space methods is the eventual drift of the end group. Maintenance of the end group at its initial position by occasional reset is performed in a way that is consistent with the overall methodology and minimally disruptive to the current conformation. The implementation permitted multiple rigid regions within the chain, enabling the simulation of domain movements where domains are rigid bodies connected by flexible interdomain regions. The method was tested on polyalanine, polyglycine, loop 6 of triosephosphate isomerase, and glutamine binding protein. Simulations of glutamine binding protein, where only 11 of the 226 residues at the interdomain bending regions were flexible, accurately reproduced the experimentally determined domain movement