Exporting to imzML Format

METASPACE requires data in the imzML format — specifically, centroided (peak-picked) data stored as two files:

• .imzML — XML file containing metadata and pixel coordinates
• .ibd — binary file containing the spectral data

Both files must be present for a valid submission. If your configuration is not listed below, contact your instrument vendor for conversion instructions. For additional help, reach out to contact@metaspace2020.org.

Supported Configurations

Vendor	Platform	Required Files	Software
Generic Multi-Vendor	Any (single file)	Any vendor format	msConvert (ProteoWizard 3.0.6938) + imzMLConverter v1.1.0
Generic Multi-Vendor	Any (multiple files)	Any vendor format	msConvert (ProteoWizard 3.0.6938) + imzMLConverter v1.1.0
Thermo	QExactive	.raw	ImageQuest v1.1.0
Thermo	QExactive	.raw	RawToImzmlConverter v1.1.4.5
Bruker	Solarix FT-ICR	.mis	SCiLS Lab 2016b+
Bruker	Solarix FT-ICR	.mis + peaks.sqlite	SCiLS Lab 2016b+
Bruker	timsTOF fleX	.mis + .tsf	SCiLS Lab 2020a

0.1 — Generic Multi-Vendor (Single Input File)

Software: msConvert GUI (ProteoWizard 3.0.6938), imzMLConverter v1.1.0

Notes:

• Applies to rectangular images where the entire dataset is contained in a single file. For datasets where each row is a separate file, see 0.2.
• This two-step conversion does not always preserve all imaging MS metadata. Where available, use a vendor-specific method instead.
• The msConvert command-line interface is also available and can perform the equivalent steps — see the msConvert documentation.

Steps:

1. Open msConvert GUI
2. Add your file: Browse → select file → OK → Add
3. Set output options:
   • Format: mzML
   • Precision: 32-bit (64-bit also works but produces larger files)
   • Choose an output directory
4. Add the Peak Picking filter:
   • Select Peak Picking from the filter drop-down
   • Check Prefer Vendor
   • Click Add
   • Ensure Peak Picking is at the top of the filter list
5. Click Start
6. Launch imzMLConverter (from run.bat)
7. Click Add and select the .mzML file produced in step 5
8. Set File organisation to Image per file
9. Enter the image dimensions (Pixels in X, Pixels in Y)
10. Add any additional metadata using the tabs at the top
11. Click Convert and choose a destination

0.2 — Generic Multi-Vendor (Multiple Files)

Software: msConvert GUI (ProteoWizard 3.0.6938), imzMLConverter v1.1.0

Notes:

• Use this method when each row of the image is stored as a separate file.
• See the notes from 0.1 for general guidance.

Steps:

1. Follow steps 1–5 from 0.1 to convert each .raw file to .mzML
2. Launch imzMLConverter (from run.bat)
3. Click Add and select all the .mzML files produced
4. Set File organisation to Row per file
5. Enter the image dimensions (Pixels in X, Pixels in Y)
6. Add any additional metadata using the tabs at the top
7. Click Convert and choose a destination

1.1 — Thermo QExactive (ImageQuest)

Software: Thermo ImageQuest v1.1.0

Notes:

• Requires a .raw file and either a .MALDIPos or .udp spatial file.
• In some cases the exported image may appear incomplete or distorted. If this occurs, try an alternative conversion method or inspect the metadata file.

Steps:

1. Open your dataset in ImageQuest
2. Go to File → Export → imzML
3. Check Export centroids only
4. Select 32-bit float precision (64-bit also works but produces larger files)
5. Click More details to enter additional experiment metadata
6. Click OK and choose a destination

1.2 — Thermo QExactive (RawToImzmlConverter)

Software: RawToImzmlConverter v1.1.4.5 (University of Giessen)

Notes:

• All image acquisition parameters must be entered manually — ensure they are recorded carefully during data acquisition.
• This software is designed and maintained for AP-SMALDI series instruments.

Steps:

1. In the Conversion tab:
   • Select your .raw imaging MS dataset
   • Check Force Centroid?
2. In the Imaging tab:
   • Verify that Dimensions (pixels in X and Y) and Pixel Size (X and Y spacing) are correctly entered
   • Confirm the scan pattern matches your experiment
3. Fill in all metadata in the remaining tabs
4. Click Convert!

2.1 — Bruker Solarix FT-ICR (flexImaging)

Software: flexImaging v3

This configuration is not currently supported. If you need to export Bruker FTMS data and do not have access to SCiLS Lab 2016b, contact contact@metaspace2020.org.

2.2 — Bruker Solarix FT-ICR (SCiLS Lab, Profile Data)

Software: SCiLS Lab 2016b or later

Notes:

• This method is still undergoing testing. Contact contact@metaspace2020.org if you encounter any issues.
• Refer to Section Exporting spectra from regions to METASPACE in the SCiLS Lab Manual, or contact SCiLS Lab Support at support.scils@bruker.com.

Steps:

1. Import the dataset in SCiLS Lab: splash screen → New → follow the dialogue
2. Open the imported dataset
3. Create a peak list before exporting:
   • Create it within SCiLS Lab using Feature Finding (T-ReX) (see SCiLS Lab Manual, Section 5.3.2), or import an externally created list (see Section 4.6)
   • Ensure the peak interval width is set to a sensible value (see Section 6.2: Accessing File Specific Properties)
4. In the Objects tab, click the export icon next to the region to export, then select Export to imzML
5. In the Export Spectra dialog:
   • Select your desired Normalization
   • Choose between the complete spectra or a reduced feature list (see SCiLS Lab Manual for details)
   • Confirm the correct polarity — this is required for METASPACE upload
   • Select a save location for the .imzML and .ibd files
6. The export runs as a background task. Progress is shown in the Tasks tab.

2.3 — Bruker Solarix FT-ICR (SCiLS Lab, peaks.sqlite)

Software: SCiLS Lab 2016b or later

Notes:

• Use this method when the dataset was acquired with on-the-fly centroid detection, indicated by a peaks.sqlite file inside the .d folder.
• Refer to Section 7.6: Exporting spectra from regions to imzML in the SCiLS Lab Manual, or contact support.scils@bruker.com.

Steps:

1. Import the dataset in SCiLS Lab: splash screen → New → follow the dialogue
2. Open the imported dataset
3. Create a peak list before exporting (see step 3 in 2.2 for details)
4. In the Objects tab, click the export icon next to the region to export, then select Export to imzML
5. In the Export Spectra dialog:
   • Select your desired Normalization
   • Choose between the complete spectra or a reduced feature list
   • Confirm the correct polarity
   • Select a save location for the .imzML and .ibd files
6. The export runs as a background task. Progress is shown in the Tasks tab.

2.4 — Bruker timsTOF fleX (SCiLS Lab)

Software: SCiLS Lab 2020a

Notes:

• METASPACE is currently incompatible with TIMS measurements.
• This method is still undergoing testing. Contact contact@metaspace2020.eu if you encounter any issues.
• Refer to Section Exporting spectra from regions to METASPACE in the SCiLS Lab Manual, or contact support.scils@bruker.com.

Steps:

1. In the Objects tab, click the export icon next to the region to export, then select Export to imzML
2. In the Export Spectra dialog:
   • Select your desired Normalization
   • Choose between the complete spectra or a reduced feature list
   • Confirm the correct polarity
   • Select a save location for the .imzML and .ibd files
3. The export runs as a background task. Progress is shown in the Tasks tab.

Software & References

Tool	Link
ProteoWizard (msConvert)	proteowizard.sourceforge.net
imzMLConverter 1.3.0 (GUI)	cs.bham.ac.uk/~ibs/imzMLConverter
jimzMLConverter 2.1.0 (CLI)	github.com/AlanRace/imzMLConverter
pyimzML	github.com/alexandrovteam/pyimzML
pyMS	github.com/alexandrovteam/pyMS
imzML format overview	maldi-msi.org