README.md

Useful tools for Colvars simulations

This directory contains both standalone tools and Colvars scripts.

File name	Summary
abf_integrate	Post-process gradient files produced by ABF and related methods, to generate a PMF. Superseded by builtin integration for dimensions 2 and 3, still needed for higher-dimension PMFs. Build using the provided Makefile.
noe_to_colvars.py	Parse an X-PLOR style list of assign commands for NOE restraints.
plot_colvars_traj.py	Select variables from a Colvars trajectory file and optionally plot them as a 1D graph as a function of time or of one of the variables.
quaternion2rmatrix.tcl	As the name says.
test_scripted_gradients.tcl	When implementing colvars as scripted functions of components, use this to test numerically the correctness of the analytical gradient.
extract_weights_biases.py	Script to read the weights and biases from a trained dense neural network model, and output them to plain text files suitable for the NeuralNetwork CV. Examples of how to train those models and save them on-the-fly can be found in the Supporting Information of this paper

File name	Summary
abmd.tcl	Adiabatic Biased MD after Marchi & Ballone JCP 1999 / Paci & Karplus JMB 1999; implemented in 20 lines of Tcl.
pathCV.tcl	Path-based collective variables after Branduardi et al. (2007). Optimized implementation that calculates only the necessary distances.