Umbrella sampling first frame histogram center distance smaller than the coresponding frame COM val

GROMACS version:
GROMACS modification: Yes/No
Here post your question
Hello guys, I’ve been running umbrella sampling. I just discovered that the center of the histogram of the first frame (d =2.624) is smaller than the coresponding COM value of the frame(d= 2.975). Is it normal? I know for other frames this could occur because the attraction force between the binding partners should conteract the umbrella force and making it deviate from the initial location. But I just don’t understand how it could be that the first frame’s histogram center from the MD simulation is smaller than the COM of that frame, which means the binding partners would be squeezed into each other should such thing happen. So does anyone know what is going on?
image

Just want this post up. So, does anyone know what was going on?

I think it’s difficult to say without more details. I think the mdp files of the pulling and umbrella sampling simulations (the first one should be enough) might help.