A super fun exercise if you're new to metabolomics is to go ahead and search your experimental data against the wrong database. Working on human metabolomics? Try using a library for pesticides or archaea or from the fossil fuel industry or something. If you're using MS1 for identifications, your number of IDs may not even decrease, particularly if you're using resolution less than 30,000! And due to the similarity of many small molecules you'll find a lot of positive hits with MS/MS as well.
In a more realistic experiment, there are MS/MS libraries for small molecules all over the place. If you get multiple conflicting hits for the same molecule from different databases, what do you do? (It's common to search your data against multiple libraries -- and multiple conflicting matches is extremely likely). Interpreting which one is correct is often a nightmare because there is no format unifier.
BOOM!
This online interface can directly import spectra from 6 different metabolomics databases and compare them. The format is fantastic -- it looks similar like the IPSA program from Coon lab for annotating peptides -- it is intuitive and the graphs are pretty and can be directly exported in a several different formats, from Support Vector Graphics through JSONs and less weird stuff in-between.
No comments:
Post a Comment