Encode percental and fold changes in simulation experiments

[matthiaskoenig]

Often in postprocessing the data is normalized on a certain time point in time courses. I.e. the percent different is plotted. E.g. glucose/DeVane1974_Fig2.

Somehow it should be simple to do generic processing of data like

ins_change = ins/ins[index=100]

This is also important for fold changes against baseline which have similar syntax.