Dear Dhiraj

I am not sure, but what you have probably done was untargeted high-resolution LC-MS metabolomics. In general, it is a very complex technique that is not easy to cover by even numerous single answers. It will be much better to get a more-less holistic vision of this approach.

The annotation of metabolites in untargeted metabolomics is very often automatic and very tentative. Again, I do not know details about your analysis, but it looks for me that you screened your data against some public libraries. In this case if the library does not have KEGG ID it will not assign it. But the problem is deeper. The automatic approach is often used and I like it too, but it has to be used with clear understanding of its limitations. Every ion you detected even with high resolution can come from several different compounds. And even if two ions do not have the same mass, they can be detected as the same masses ions because of very small, but still existing instrumental error, or just simply be isomers.

In summary, if you indeed used untargeted metabolomics, annotation of metabolites is not the first thing to focus on (in my opinion). If you really need to check if among all these compounds there are several specific metabolites, run their standards, as automatic annotation will only provide you match with the library based on exact mass, which can be achieved by combination of different elements. RT on LC-MS is also not an easy thing. If you have your own library it can work well, but it is very difficult to replicate exact conditions from somebody's else, and this will lead to RT shifts.

Thank you Albert. It's really a very much useful explanation. Yes, I am dealing with an untargeted approach. Before MS2 I want to have a clear idea of my compounds to validate them by MS2 fragmentation.

Hi. I would like to mention that: 1) some annotated compounds are contaminations from the sampling equipment or GC/LC-MS system and thus you will not be able to find them KEGG database, 2) if the annotation was made only on the basis of MS spectrum the score of identification is important. You can identify the daughter fragment but not the parent compound.