Program

Poster Presentations

Day 4, May 18（Fri.）　　Poster

4P-36　　PDF

Metadata Curation for fully utilizing raw MS data in jPOST repository

(¹Kumamoto Univ., ²Niigata Univ., ³DBCLS, ⁴Niigata Univ., ⁵Kyushu Univ., ⁶Kyoto Univ., ⁷Kyoto Univ., ⁸Trans-IT)
^oDaiki Kobayashi¹, Norie Araki¹, Shujiro Okuda², Yu Watanabe², Yuki Moriya³, Shin Kawano³, Tadashi Yamamoto⁴, Masaki Matsumoto⁵, Tomoyo Takami⁵, Akiyasu Yoshizawa⁶, Tsuyoshi Tabata⁶, Mio Iwasaki⁷, Naoyuki Sugiyama⁶, Satoshi Tanaka⁸, Susumu Goto³, Yasushi Ishihama⁶

A large amount of mass spectrometry (MS)-based proteomics data has been generated owing to the recent striking advance in MS instruments and methodology. The construction of management system for these MS files, such as a raw data repository, a standardized re-analysis protocol, and integrated database, is much helpful to publicly utilize these data sets. Here we have initiated the jPOST (Japan ProteOme STandard Repository/Database) project1 to construct such kind of data management system. To fully utilize these MS datasets, the metadata annotation of the MS raw file is important to precisely identify the sample sources, the sample preparation methods, and the MS acquisition settings. However, these metadata are sometimes not in detail, or have inappropriate annotations. First, these files have been divided into each reanalysis unit according to these metadata annotation because raw MS files are re-analyzed to identify peptides/proteins with high quality. Then, these metadata has been converted to the controlled vocabularies used in jPOST database. Especially, we have established the classification method of disease-related samples based on sample types, organs, and disease types. In this presentation, we show some cases of the curation workflow for the metadata annotating jPOST MS files.