MS-DIAL project

Program download

Main program is available from here. MS-DIAL program.

MS-DIAL tutorial is downloadable from here. MS-DIAL tutorial.

File converter is downloadable from here. File converter.

FAQ for MS-DIAL and ABF file converter.

MS-DIAL mathematics is downloadable from here. MS-DIAL mathematics.

Please cite

- H.Tsugawa et al (2015) MS-DIAL: data independent MS/MS deconvolution for comprehensive metabolome analysis. Nature Methods, 12, 523-526 [PubMed]


See FAQ web page too, for the usage of MS-DIAL and ABF file converter.

Demonstration files

LC/MS/MS data set (.wiff and .wiff.scan, and the converted abf files) for lipid profiling of algae is available from here. data independent acquisition (SWATH) and data dependent acquisition (IDA) for algae lipidomics; Also see MSDIAL quick start

LC/MS/MS data set (.abf, data dependent MS/MS acquisition) for hydrophilic metabolome profiling of Wine is available from here. data dependent acquisition for Wine metabolome;

GC/MS data set (.cdf and .abf files) for small biomolecule analysis of Arabidopsis thaliana and Alage is available from here. GC/MS data set;

HILIC-SWATH-MS data (.wiff, .wiff.scan, and .abf) that we used for the explanation of mass spectral deconvolution is downloadable from here. HILIC(+)-SWATH-MS

Curated spectra download (MSP format: Last edited in Sep.16, 2016)

All MassBank, MoNA, Respect, and GNPS: Positive (14,951 records), Negative (6,819 records), and All (21,770 records)

MassBank MS/MS Positive (8412 records) MassBank (Positive).

MassBank MS/MS Negative (5038 records) MassBank (Negative).

ReSpect MS/MS Positive (2768 records) ReSpect (Positive).

ReSpect MS/MS Negative (1628 records) ReSpect (Negative).

GNPS MS/MS Positive (3771 records) GNPS (Positive).

GNPS MS/MS Negative (153 records) GNPS (Negative).

All MassBank and MoNA RT, RI, and EI (15,302 records) All of public records.

Fiehn BinBase DB (Rtx5-Sil MS, FAMEs RI, 1021 records) Fiehn BinBase.

Fiehn BinBase DB (Rtx5-Sil MS, predicted Kovats RI, 1021 records) Fiehn BinBase.

RIKEN DB (Rtx5-Sil MS, Kovats RI, 241 records) RIKEN.

Kazusa DB (Rtx5-Sil MS, Kovats RI, 273 records) Kazusa.

GL-Science DB (InertCap 5MS-NP, Kovats RI, 494 records) GL-Science.

Osaka Univ. DB (CPSil-8CB, Kovats RI, 430 records) Osaka Univ..

LipidBlast fork (Last edited in Feb.4, 2017)

Currently, MS-DIAL internally has in silico MS/MS spectra for lipid identifications. The below is the LipidBlast (fork) templates that MS-DIAL partially uses.

LipidBlast template for glycerolipids.

LipidBlast template for sphingolipids.

The original LipidBlast is available from here.


This project was supported by NSF-JST Strategic International Collaborative Research Program (SICORP) for JP-US Metabolomics.
MS-DIAL is mainly developed between UC Davis Prof. Oliver Fiehn team and RIKEN CSRS (and NIG) Prof. Masanori Arita team.

Main developer: Hiroshi Tsugawa (RIKEN)
Main contributor: Diego Pedrosa (UC Davis)
Other supporters: Haruki Uchino (RIKEN, Keio Univ.) and Gert Wohlgemuth (UC Davis)


MS-DIAL was launched as a universal program for untargeted metabolomics that supports multiple instruments (GC/MS, GC/MS/MS, LC/MS, and LC/MS/MS) and MS vendors (Agilent, Bruker, LECO, Sciex, Shimadzu, Thermo, and Waters). Common data formats such as netCDF (AIA) and mzML, can also be managed in our project. In addition, we released several MSP files including both EI- and MS/MS spectra as a ‘start-up kit’. Moreover, MS-DIAL internally has a version of Fiehn lab’s GC/MS database (oriented by FAME RI index), and in silico retention time- and MS/MS database for LC/MS/MS based lipidomics. The isotope labeled tracking can also be executed in LC/MS project. It features (1) spectral deconvolution for both GC/MS and data-independent MS/MS, (2) streamlined criteria for peak identification, (3) support of all data processing steps from raw data import to statistical analysis, and (4) user-friendly graphic user interface.

Overview of MS-DIAL software

1. Convert vendor's format file into ABF format file

MS-DIAL imports our common data format (ABF). The file converter can be freely downloaded from Download
Currently, all of major MS vendor's format as well as common data format such as mzML and netCDF is supported.

2. Start up of a MS-DIAL project

MS-DIAL provides the data processing solution for all type of data sets including GC/MS, data dependent/independent acquisition, positive/negative ion mode, and metabolomics/lipidomics application. In the start up window, users can choose (A) project type: GC/MS or LC/MS, (B) data type: centroid or profile, (C) ion mode: positive or negative, and (D) omics type: metabolomics or lipidomics.

3. Parameter settings

Some parameters should be determined for data collection, peak detection, de-convolution, identification, and alignment. See the tutorial and mathematics for the detail. As the starter kit, you can utilize the above NIST MSP DBs. Also, the lipid identification is now straightforward since MS-DIAL internally has in silico retention time and MS/MS spectra of major lipid species.

4. Main window (data curation, normalization, statistical analysis and export)

Graphical user interface of MS-DIAL is optimized for metabolomics and lipidomics. The user can easily confirm the identification or peak alignment result and manually curate the identification result.

5. Futher statistics

MS-DIAL supports interpolation methods for missing values, normalization methods (internal standard, LOESS/Cubic-spline etc.), and principal component analysis.
Moreover, we also provide a tool for statistics by microsoft excel at excel macro based statistical analysis tool page.

Source code

The main source code is available from here. MS2Dec sample program.
The demo file (mzML: HILIC-LC/SWATH/MS for a human plasma) and an Analyst experiment file (25-Da setting) are downloadable from here. MS2Dec sample program demo file.

This program is for one mzML file of SWATH-MS exported by ProteoWizard (64 bit and profile mode).
This program exports the peak detection and deconvolution result, i.e. retention time, precursor m/z, abundance, raw MS/MS spectrum, and deconvoluted MS/MS spectrum.

RIKEN Center for Sustainable Resource Science (formerly Plant Science Center)
Practical and Useful Data Analysis Tools for Metabolomics Research