From user feedback:
for graphML files "the vast majority of input columns are not retained, including the original node name column(!). This means the only feature left to relate back to the original network is the parent mass. This makes the clusters very difficult to relate back to the original input. Is there some way of retaining a unique identifier (or even better, all of the original input columns, like retention time) so that the input and output can be more easily collated?"
From user feedback:
for graphML files "the vast majority of input columns are not retained, including the original node name column(!). This means the only feature left to relate back to the original network is the parent mass. This makes the clusters very difficult to relate back to the original input. Is there some way of retaining a unique identifier (or even better, all of the original input columns, like retention time) so that the input and output can be more easily collated?"