SE133:/S2/M1/D1

Sample Set Information

ID	TSE7
Title	Regular expressions of MS/MS spectra for partial annotation of metabolite features
Description	Partial annotation and characterization of metabolite structures on the basis of data from tandem mass spectrometry (MS/MS) spectra are technical bottlenecks in metabolomics. Novel approaches should be explored for evaluation of spectral similarities among structurally related compounds as well as for description of fragmentation motifs commonly observed in MS/MS spectra.
Authors	Fumio Matsuda
Reference	Matsuda (2016) Metabolomics, July, 12:113
Comment	MS/MS strings of MassBank dataset and MS/MS strings of Arabidopsis (ATH) and rice (OSA) MS/MS spectra data are stored in DROP Met as "Test dataset for the regular expression of MS/MS spectra data"

doi: 10.1007/s11306-016-1052-3

The raw data files are available at DROP Met web site in PRIMe database of RIKEN.

Sample Information

ID	S2
Title	Rice
Organism - Scientific Name	Oryza sativa
Organism - ID	NCBI taxonomy 4530
Compound - ID
Compound - Source
Preparation	The plant population consisted of 85 back‐crossed inbred lines derived from the cross Sasanishiki/Habataki//Sasanishiki///Sasanishiki (Sasanishiki x Habataki) (Nagata et al., 2002b). Seeds from the experimental lines were grown in a paddy field at the National Institute of Agrobiological Sciences (Tsukuba, Japan) in 2005 and 2007, employing similar cultivation schedules. The seeds of the 2005 and 2007 harvests were used for metabolome analysis. One hundred dehulled seeds obtained from whole seeds harvested from 10 independent plants were ground to a fine powder using an MM300 mixer mill (Retsch, http://www.retsch.com/) at 20Hz for 2min in a stainless steel grinding vessel. The powder was divided between small sample tubes (50–100mg) under nitrogen, and the samples were stored at −80°C until analysis.
Sample Preparation Details ID
Comment	Matsuda, F., et al. (2012). Dissection of genotype-phenotype associations in rice grains using metabolome quantitative trait loci analysis. The Plant Journal, 70, 624–636.

PubMed

Analytical Method Information

ID	M1
Title	LC-QTOF-MS
Method Details ID	MS2
Sample Amount	3 μL
Comment

Analytical Method Details Information

ID	MS2
Title	LC-QTOF-MS/MS Method (rice)
Instrument	LC, Waters Acquity UPLC system; Q-TOF-MS, Waters Q-Tof Premier
Instrument Type	UPLC-QTOF-MS
Ionization	ESI
Ion Mode	Positive
Description	Sample metabolomic data were acquired using four time-of-flight mass spectrometers (TOF-MS), including CE-TOF-MS for analysis of polar metabolites, GC-TOF-MS for analysis of primary metabolites, LC-Q-TOF-MS for analysis of secondary metabolites and LC-IT-TOF-MS for analysis of lipids. Analysis was performed in triplicate, and the samples were extracted from each pipeline using optimized methods. For the LC-Q-TOF -MS pipeline, 100 mg of rice seed powder was homogenized in 10 volumes of 5% aqueous methanol containing 0.1% acetic acid (Tokyo Kasei, http://www.tciamerica.com/) using an MM300 mixer mill (Retsch, http://www.retsch.com/) with a zirconia bead for 10 min at 20 Hz. Next, the samples were centrifuged at 15000 g for 10 min. The supernatant (500 ll) was diluted to 5.0 ml using water containing 0.1% acetic acid, and was applied to a 3 cc OASIS HLB cartridge (Waters, http://www .waters.com/) that had been equilibrated with 0.5% aqueous methanol containing 0.1% acetic acid. After pre-equilibrium with 5 ml of 0.5% aqueous methanol containing 0.1% acetic acid, metabolites were eluted with 3 ml of 90% aqueous methanol containing 0.1% acetic acid. The eluate was dried under vacuum and then suspended in 100 ll water containing the internal standard (0.5 mg/ml lidocaine). Insoluble residue was removed by filtration using an Ultrafree-MC filter with 0.2 lm pore size (Millipore, http://www.millipore.com/). The samples (3µl) were subsequently subjected to metabolome analysis by liquid chromatography coupled with electrospray quadrupole time-of-flight tandem mass spectrometry (LC-ESI-Q-TOF-MS) using an Acquity BEH ODS column (LC, Waters Acquity UPLC system; Q-TOF-MS, Waters Q-Tof Premier).
Comment_of_details

PubMed

Data Analysis Information

ID	D1
Title	Data analysis
Data Analysis Details ID	DS1
Recommended decimal places of m/z
Comment

Data Analysis Details Information

ID	DS1
Title	Methods
Description	2.1 MS/MS spectra MassBank data with fragment assignment information (Fragment ion and neutral loss matrix) were obtained from Metabolime.jp (http://metabolomics.jp/wiki/Index:MassBank) (Arita and Suwa 2008; Horai et al. 2010). All procedures were based on in-house scripts written in Perl 5.12. Similarity networks were visualized by Cytoscape 3.2.1 (Shannon et al. 2003). All MS/MS spectral libraries obtained from various tissues of Arabidopsis (ATH06p–ATH014p) (Matsuda et al. 2010) and rice cultivars (OSA06p–OSA014p) (Matsuda et al. 2012) were downloaded from the PRIME webpage (http://prime.psc.riken.jp/, Supplementary Table S1) (Sakurai et al. 2013). Each entry in the MS/MS spectral library has a unique ID such as ATH14p11717 (Matsuda et al. 2009). For visualization and analysis of these MS/MS spectra, (1) please go to the MS2T section of the RIKEN PRIMe web page (http://prime.psc.riken.jp/lcms/ms2tview/ms2tview.html), (2) input IDs (e.g., ATH14p11717) into the “MS/MS spectrum tag(MS2T) accession codes” field, and (3) press the Start button. (4) A list of MS/MS spectra will appear. (5) Press the Detail:[ATH14p11717] button for detailed analysis of the MS/MS spectra. The “Molecular formula search” function is also available. 2.2 Conversion of MS/MS spectra to MS/MS strings MS/MS strings were generated from the MS/MS spectra by the following procedure. For each entry in the MS/MS spectral libraries, 1. The entry was discarded when intensity of the most intense fragment (base peak) was less than 30 counts per second. 2. The fragments whose intensity was less than that of the 12th most intense signal were ignored. Fragment signals of weak intensity (less than 1.0 % of the base peak) were also ignored. 3. The chemical formula of precursors was assigned according to the seven golden rules (Kind and Fiehn 2007). 3.1 The chemical formulas including CHONS atoms with 0 ≤ N ≤ 5 and 0 ≤ S ≤ 3 were considered. All formulas within the threshold (10 mDa [m/z < 500] and 20 ppm [m/z > 500]) were used to construct MS/MS strings. 3.2 For each chemical formula of a precursor, the chemical formulas of all fragments were deduced according to the seven golden rules, with the threshold at 20 ppm. The numbers of atoms in a molecular fragment had to be less than those in the precursor. 3.3 The total intensity of fragment signals whose chemical formulas were successfully elucidated (I selected) was calculated. The chemical formula of a precursor was accepted when I selected ÷ I total was greater than 0.8, where I total indicates total intensity of all fragment signals (Meringer and Schymanski 2013). 4. An entry without an accepted chemical formula of a precursor was discarded. 5. All possible MS/MS strings were generated using the following format: [formula of neutral loss 1]:[formula of fragment 1];[formula of neutral loss 2]:[formula of fragment 2];… [formula of neutral loss n]:[formula of fragment n];. 5.1 The threshold for finding a chemical formula of a neutral loss was 15 mDa. 5.2 One neutral loss involving negative atom numbers (such as; C1H-2:) was allowed in 1 MS/MS string. 5.3 One fragment with an integer index of hydrogen deficiency (IHD) was allowed in 1 MS/MS string. The MS/MS strings that were generated by this procedure were incomplete because steps 5.2 and 5.3 are heuristic conditions designed to restrict a combinatorial explosion of the number of MS/MS strings. All MS/MS string libraries derived from MassBank, Arabidopsis, and rice data were downloaded from the project homepage (http://www-shimizu.ist.osaka-u.ac.jp/hp/en/software/regex.html) and DROP Met in PRIMe website (http://prime.psc.riken.jp/?action=drop_index).
Comment_of_details	MS/MS strings of MassBank dataset (Supplementary Data 1) and MS/MS strings of Arabidopsis (ATH) and rice (OSA) MS/MS spectra data. http://prime.psc.riken.jp/archives/data/DropMet/022/

The raw data files are available at DROP Met web site in PRIMe database of RIKEN.

The peak data files of this analysis are available at MS2T.