Abstract
The plant metabolome is extremely diverse comprising several thousands of chemical compounds. Current approaches for profiling the metabolome are hampered by the fact that no analytical platform can give comprehensive coverage of all biological metabolites. A promising way to work around this problem is to use several complementary analytical platforms in parallel and combine the resulting data sets. However, stitching together data from different platforms poses several theoretical and practical challenges for how to best normalize and summarize the data to obtain a consensus data set. Here we present a novel strategy for solving this task and a freely available software tool that implements our ideas. To validate our method we tested it on an experimental data set obtained using both GC-, and CE-MS and show its performance in comparison with alternative approaches.
