Navigating common pitfalls in metabolite identification and metabolomics bioinformatics

Metabolomics, the systematic analysis of small molecules in a given biological system, emerged as a powerful tool for different research questions. Newer, better, and faster methods have increased the coverage of metabolites that can be detected and identified in a shorter amount of time, generating highly dense datasets. While technology for metabolomics is still advancing, another rapidly growing field is metabolomics data analysis including metabolite identification. Within the next years, there will be a high demand for bioinformaticians and data scientists capable of analyzing metabolomics data as well as chemists capable of using in-silico tools for metabolite identification. However, metabolomics is often not included in bioinformatics curricula, nor does analytical chemistry address the challenges associated with advanced in-silico tools.

In this educational review, we briefly summarize some key concepts and pitfalls we have encountered in a collaboration between a bioinformatician (originally not trained for metabolomics) and an analytical chemist. We identified that many misunderstandings arise from differences in knowledge about metabolite annotation and identification, and the proper use of bioinformatics approaches for these tasks. We hope that this article helps other bioinformaticians (as well as other scientists) entering the field of metabolomics bioinformatics, especially for metabolite identification, to quickly learn the necessary concepts for a successful collaboration with analytical chemists.

We summarize important concepts related to LC-MS/MS based non-targeted metabolomics and compare them with other data types bioinformaticians are potentially familiar with. Drawing these parallels will help foster the learning of key aspects of metabolomics.