'Reviewing the literature, the paper suggests that over 70% of the putative "dark metabolome" is merely an experimental artifact. This is of course not good news for all the people who have been trying to assign these peaks, nor for the other groups who have been trying to model metabolism computationally based on the typical mass spec assumption that any unassigned peaks are due to separate new species. People are going to have to rethink a lot of the ESI mass spec data out there, it seems.'
https://www.science.org/content/blog-post/phantom-metabolites