Fig. 1

Overview of the method for automated identification of fragment formulae. Orange box (top): input measured data. Two mauve boxes (left): input chemical data [56]. Yellow boxes are steps done just once. Steps 1 to 4: steps of initialisation. Green boxes, steps 5 to 8: steps repeated until a certain fraction of the measured signal has been reconstructed, here 95%. Steps 9 and 10, yellow box: final steps, done just once. Blue box: output data, list of most likely fragments together with associated likelihood and ranking