Data mining of natural hazard biomarkers and metabolites with integrated metabolomic tools.

Abstract Data mining was one of the most important challenges in natural product analysis and biomarker discovery. In this work, we proposed an integrated data analysis protocol for natural products annotation and identification in data–dependent acquisition. Firstly, natural products and structure–related compounds could be identified by comparing mass spectrum behavior with commercial standard. Secondly, diagnostic fragmentation filtering (DFF) function in MZmine (http://mzmine.github.io/) was investigated for screening specific conjugation compounds with the same neutral loss. Thirdly, we present feature–based molecular networking (FBMN) in GNPS (https://gnps.ucsd.edu/) as a chromatographic feature detection and alignment tool. In addition, FBMN could enable natural products analysis based on molecular networks. This proposed integrated protocol should facilitate metabolomic data mining and biomarker discovery.