Home

Thyra

Thyra (from Greek thyra, meaning "door" or "portal") converts Mass Spectrometry Imaging (MSI) data into the standardized SpatialData/Zarr format -- your portal to spatial omics analysis workflows.

Why Thyra?¶

Mass spectrometry imaging produces rich spatial-molecular data, but every vendor stores it differently. Downstream tools -- napari, squidpy, scanpy -- expect a common format. Thyra bridges that gap:

 .imzML  ──┐                        ┌── napari visualisation
 .d      ──┼──  thyra  ──> .zarr  ──┼── squidpy / scanpy analysis
 .raw    ──┘   (SpatialData)        └── custom Python workflows

The output is a single SpatialData/Zarr directory containing intensity matrices, TIC images, optical images, pixel geometries, and full metadata -- ready for any tool in the scverse ecosystem.

Features¶

	Feature	Description
Formats	Multiple inputs	ImzML, Bruker (.d timsTOF + Rapiflex), Waters (.raw directory), PHI SmartSoft-TOF (.raw file)
Output	SpatialData/Zarr	Cloud-ready, chunked, standardised
Scale	Memory efficient	Streaming mode for 100+ GB datasets
Optics	Optical alignment	Automatic MSI-to-microscopy registration (Bruker)
Regions	Multi-region	Handles slides with multiple tissue sections
Resampling	Physics-aware	Instrument-specific mass axis resampling (on by default in the CLI, opt-in from the Python API)
3D	Volume support	Process as 3D volume or separate 2D slices
Platform	Cross-platform	Windows, macOS, Linux

Quick Start¶

Install¶

pip install thyra

Convert¶

CLIPython

thyra input.imzML output.zarr

from thyra import convert_msi

success = convert_msi("input.imzML", "output.zarr")

Explore the output¶

import spatialdata as sd

sdata = sd.read_zarr("output.zarr")

# Intensity matrix (pixels x m/z bins)
table = sdata.tables["msi_dataset_z0"]
print(f"Shape: {table.shape}")
print(f"m/z range: {table.var['mz'].min():.1f} -- {table.var['mz'].max():.1f}")

# TIC image
import numpy as np
tic = np.asarray(sdata.images["msi_dataset_z0_tic"])[0]

What is in the output?

See Output Format for the full structure: tables, TIC images, optical images, pixel shapes, regions, and metadata.

Supported Formats¶

Input¶

Format	Path	Instruments
ImzML	`.imzML` file	Any vendor exporting to the open standard
Bruker	`.d` directory	timsTOF fleX, Rapiflex MALDI-TOF
Waters	`.raw` directory	MassLynx imaging (DESI, MALDI)
PHI	`.raw` file	SmartSoft-TOF nanoTOF (ToF-SIMS)

.raw is claimed by two vendors and resolved by shape: Waters writes a directory, PHI writes a single file. See Supported Formats.

Output¶

Format	Description
SpatialData/Zarr	The scverse standard for spatial omics -- cloud-ready, chunked, with coordinate transforms

Next Steps¶

Getting Started -- installation, first conversion, common workflows
Tutorial -- step-by-step walkthrough, from an example dataset to ion images
Supported Formats -- every input format, how it is detected, what metadata it supplies
CLI Reference -- every command-line option explained
Resampling -- how the common mass axis is chosen, and how to control it
Output Format -- what the .zarr contains and how to use it
API Reference -- Python API documentation