Weighted histogram

GROMACS version: 2020
GROMACS modification: No

Dear all,
Concerning the gmx clustsize, below I have two questions on which you comments are highly appreciated;

(For a single frame)
gmx clustsize \
-f conf.gro -s conf.tpr -nc nclust.xvg \
-o csize.xpm -ow csizew.xpm \
-hc histo-clust.xvg -ac avclus.xvg \
-temp temp.xvg -cut 0.35 -mol

  1. I wonder if the “-hc histo-clust.xvg (or \frac{\text{histo-clust.xvg}}{\text{Total number of clusters}})” gives a weighted histogram?

  2. How does Gromacs calculate the average cluster size (-ac avclus.xvg) for a frame?
    The average value of a histogram is calculated as following
    \frac{\sum\limits_{i=1}^{\textrm{max~size}}~(\text{(cluster size)}_i \times \text{(number of cluster with size}_i) )}{\sum\limits_{i=1}^{\textrm{max~size}}~\text{(number of cluster with size}_i)}
    From “-hc histo-clust.xvg” for a sample frame, the above formula gives an average value much lower than what the “-ac avclus.xvg” has reported for the frame!

Thank you

@jalemkul @dspoel
Any comment, please?

Thank you.