1. I am processing my umbrella sampling simulations using the Weighted Histogram Analysis Method in Gromacs (gmx wham). However, I would like to report the mean dG and a standard deviation. To do this I plan on using bsProfs.xvg, which contains all 100 bootstrapping profiles (i.e. used the flag -nBootstrap 100), to derive a mean dG and a standard deviation. However, I do not wish to manually estimate the dG by inspecting the well depth of each individual profile. Is there an automated script available to do this?

2. Additionally, for greatest accuracy of the error estimate, do I need to run replicas of all umbrella sampling windows or is it sufficient to run these once, provided that the histograms are sufficiently overlapping?

Thanks so much in advance!

EDIT: I have now prepared a script to calculate mean dG and stdev (part 1.) but still I require an answer to part 2. re the requirement for replicas.