演題概要

ポスター発表

第３日　６月１９日（金）　　P会場

3P-40(3B-O1-1030)　　PDF

MS-FINDER: ケムおよびバイオインフォマティクスリソースを統合することによる未知化合物の同定戦略

(¹理研CSRS・ ²UC Davis・ ³ライフィクス・ ⁴千葉大院・ ⁵遺伝研)
^o津川裕司¹・ Kind, Tobias²・ 中林亮¹・ 行平大地³・ 斉藤和季^1,4・ Fiehn, Oliver²・ 有田正規^1,5

Liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) is the preferred tool for untargeted metabolomics: detection, identification, and quantification of biomolecules. The main bottleneck in its discovery phase is the structure estimation of unknown metabolites. Since data-independent acquisition in combination with mass spectral deconvolution provides comprehensive MS/MS spectra for all precursor ions, we propose the following three-step approach. First, formula calculation predicts a molecular formula from both MS1 and MS/MS in combination with the seven golden rules and a database for neutral losses. Second, structure data that match with the formula are downloaded via the PubChem Rest Service, and are computationally fragmented with the Chemistry Development Kit to check consistency. Finally, the likelihood scores are calculated to rank candidate structures.