It could be useful to support systematic uncertainties in these plots. The question of how is the big one.... do I ask for a TRExFitter config and workspace? how do we map the names of the samples (TRExName sample attribute?) ? Should I use informatio from uhepp (can make use of the API to retrieve systematics).