Summary statistics of selection are incorrect after projection

[puehringer]

One needs to reload the page to see the correct summary. In this example, I projected on the ligand with high weight, such that I expect the clusters to correspond to one ligand. But a selection after the projection just shows a weird pattern, only after reloading the page (full dataset reload) the selection of a cluster is a single ligand. Build upon #95

Screencast.from.2023-05-08.15.22.56.webm