Skip to content

System requirements

Installation

Installation of mzmine is described on the getting started page. mzmine is available as an installable package or a portable version. The portable version does not require administrator rights to be run, making it useful for users without elevated permissions.

Hardware requirements

Minimum (small datasets < 30 LC-MS files):

  • 64 bit CPU, 4 Cores (2.5 GHz, Intel Core i5 or AMD Ryzen 5), 16 GB RAM, secondary Sata SSD drive (512 GB), integrated graphics
  • Display 1920 x 1080 pixel (Full HD)
  • Internet connection for login and spectral library/machine learning model download
  • Keyboard and mouse

Recommended (medium to large data sets ≥ 100 LC-MS files, or IMS-MS in general):

  • 64 bit CPU, ≥ 16 Cores (≥3 GHz, Hyper threading), ≥ 64 GB RAM (scales with data files), secondary NVMe SSD drive (≥ 1 TB), integrated graphics
  • Display 2560 x 1440 pixel or higher
  • Internet connection for login and spectral library/machine learning model download
  • Keyboard and mouse
  • dedicated GPU (CUDA enabled) for accelerated machine learning models (not required)

Info

Offline login is possible, see Offline use.

Info

Processing speed scales with CPU cores/threads and speed of the SSD for temporary files. We recommend setting the temporary files directory to a fast, secondary SSD in the mzmine preferences (CTRL+P, Project → Preferences). (see )

Software requirements

  • Up-to-date operating system, e.g., Windows 10 or newer, recent Linux or MacOS (academic only) versions.
  • mzmine does not require a dedicated Java installation, as it is a self-contained Java software with its own Java Virtual Machine. All requirements are shipped with mzmine.
  • Microsoft Visual Studio C++ Redist for Bruker raw data import download page
  • MSConvert (on Windows) for native Agilent, Sciex, Waters, Shimadzu, and MOBILion data support download page

Internet connection

  • An internet connection is recommended, but not strictly required for core processing
  • Offline user login is possible, see Offline use.
  • To allow mzmine to download spectral libraries and the recent versions of machine learing models for spectral networking using MS2Deepscore and DReaMS, an internet connection is required
  • A proxy may need to be set in the Preferences if your University/Company uses one (Project -> Preferences, CTRL+P)
  • Required URLS (in case University/Company blocks unknown):
    • https://auth.mzio.io/ must be accessible user login
    • https://zenodo.org/ machine learning models
    • https://djl.ai/ machine learning models
    • https://zenodo.org/ spectral libraries
    • https://external.gnps2.org/gnpslibrary spectral libraries

Operating system compatibility

Windows

Currently, all modules are compatible with Microsoft Windows 10 and higher.

Some libraries for the raw data support for vendor-specific formats are only available for Windows. Read more about data support and data conversion.

Linux

Some libraries for the raw data support for vendor-specific formats are only available for Windows.

The Linux version supports raw data formats from:

  • Thermo, Bruker, Waters

Data from other Vendors may need to be converted to the open .mzML format before, including:

  • Agilent, Sciex, Shimadzu, MOBILion

macOS

Some libraries for the raw data support for vendor-specific formats are only available for Windows and Linux.

The macOS version supports raw data formats from:

  • Thermo

Data from other Vendors may need to be converted to the open .mzML format before, including:

  • Agilent, Sciex, Shimadzu, MOBILion, Bruker, Waters