Albicidin is produced by the sugarcane pathogenic bacterium Xanthomonas albilineans and is a potent antibiotic specifically targeting the bacterial DNA gyrase with an IC50 value in the nanomolar range. The structure of albicidin, which has remained unsolved for more than three decades since its first description, has been solved by a combination of extensive NMR and HR-MS/MS experiments. An inherent problem for the structure elucidation was the low amount of the albicidin production in cell culture. The low production forced us to cultivate several hundred liters in order to purify a few milligrams required for NMR experiments. In comparison, LC-HR-MS/MS has the great advantage of several orders of magnitude and high sensitivity. Thus, this approach requires much less material. But more importantly, purification of compounds is not necessary for structural analysis. Due to the application of targeted LC-MS/MS experiments, we were able to propose the structure of N-terminal carba- S158 moylated and b-methoxylated albicidin derivatives. Both of them are currently chemically synthesized and studied in a medicinal chemistry campaign. In order to search for more pharmaceutically relevant albicidin derivatives, we applied non-targeted LC-HR-MS/MS approaches combined with mass spectral networking, a recently introduced approach for non-targeted MS/MS data-analysis. However, as for the structural elucidation of albicidin, the even lower concentrated derivatives resulted in difficulties to obtain useful MS/ MS spectra. Since we typically operate the mass spectrometer in data-dependent acquisition mode (DDA), it inherently selects high abundant compounds of cell extracts for MS/ MS scans and possibly ignore low abundant compounds. The same goes for the network analysis which is based on similarity of MS/MS spectra and may not take potential ion species into account. To overcome this problem, we implemented a bioactivity-guided fractionation by solid phase extraction and semi-