# Mass Spectrometry Tune Data Loading

> Use when you have positive- or negative-mode tune reference compound data stored in HDF5 format (e.g., example_tune_pos.h5) and need to extract the tune mass spectrum for CCS calibration. This skill is the entry point before applying deimos.calibration.

- Skill: `holobiomicslab/mass-spectrometry-tune-data-loading` (Agent Skill)
- Install (CLI): `npx skillmds@latest add holobiomicslab/mass-spectrometry-tune-data-loading`
- Raw SKILL.md: https://api.skillmd.com/api/skills/holobiomicslab/mass-spectrometry-tune-data-loading/raw
- Safety review: PASS (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- License: CC-BY-4.0
- Author: HolobiomicsLab (https://skillmd.com/u/holobiomicslab)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/holobiomicslab/mass-spectrometry-tune-data-loading

---


# mass-spectrometry-tune-data-loading

## Summary

Load collision cross section (CCS) calibration tune reference data from HDF5 files using DEIMoS, preparing standardized tune compounds for downstream calibration workflows. This skill is essential for establishing the m/z–CCS relationship required for accurate ion mobility spectrometry analysis.

## When to use

You have positive- or negative-mode tune reference compound data stored in HDF5 format (e.g., example_tune_pos.h5) and need to extract the tune mass spectrum for CCS calibration. This skill is the entry point before applying deimos.calibration.tunemix() to establish a calibration model with known m/z and CCS values.

## When NOT to use

- Input data is already in processed feature table format (peaks detected and aligned); use deimos.load() only on raw or tune reference data.
- Data is stored in non-HDF5 formats (e.g., raw mzML, NetCDF); convert using ProteoWizard msconvert or appropriate format-specific loader first.
- You are loading sample data for feature detection rather than tune reference compounds; tune data serves a calibration-specific role and should not be used as analysis samples.

## Inputs

- HDF5 file containing tune reference data (e.g., example_tune_pos.h5)
- key parameter specifying the dataset path within the HDF5 file (string, typically 'ms1')

## Outputs

- Tune mass spectrum DataFrame with columns including m/z, intensity, and optional drift_time/retention_time dimensions
- Structured data object ready for calibration function input

## How to apply

Use deimos.load() with the HDF5 tune data file path and specify the key parameter (typically 'ms1' for mass spectrometry data). The function returns a structured DataFrame containing m/z, intensity, and other spectral dimensions for the tune reference compounds. Extract the loaded tune data object for subsequent calibration. The loaded data should span the expected m/z range of the tune mix (e.g., m/z 118.086255–1521.971475 for positive-mode compounds) and contain intensity values for all reference compound peaks.

## Related tools

- **deimos** (Provides the deimos.load() function to parse and deserialize HDF5 tune data into memory-efficient DataFrames) — https://github.com/pnnl/deimos
- **Python** (Runtime environment for executing deimos.load() and downstream calibration workflows)
- **ProteoWizard** (Used to convert tune data from other formats (e.g., native vendor formats) to mzML before HDF5 conversion if needed)

## Examples

```
tune_pos = deimos.load('example_tune_pos.h5', key='ms1')
```

## Evaluation signals

- Returned DataFrame contains expected columns: m/z, intensity, and (optionally) drift_time, retention_time, or other dimensions
- m/z values span the known range of tune reference compounds (e.g., 118.086–1521.971 for positive-mode mix)
- Intensity values are non-negative and non-zero for all reference compound peaks
- DataFrame can be passed without error to deimos.calibration.tunemix() for CCS calibration
- Row count and data shape are consistent with the number of scans/spectra in the tune file

## Limitations

- HDF5 file must be properly formatted with the specified key present; missing or malformed keys will raise KeyError
- Loading large tune datasets may require significant memory; no streaming or chunked loading option is documented
- deimos.load() is largely agnostic to instrument type but relies on HDF5 data structure conforming to DEIMoS conventions
- Tune data quality depends on upstream data acquisition and conversion; corrupted or incomplete tune files will propagate errors to calibration

## Evidence

- [results] tune_pos = deimos.load('example_tune_pos.h5', key='ms1'): "tune_pos = deimos.load('example_tune_pos.h5', key='ms1')"
- [other] Load the positive-mode tune data from example_tune_pos.h5 using deimos.load() with key 'ms1': "Load the positive-mode tune data from example_tune_pos.h5 using deimos.load() with key 'ms1'"
- [readme] DEIMoS, or Data Extraction for Integrated Multidimensional Spectrometry, is a Python application programming interface and command-line tool: "DEIMoS, or Data Extraction for Integrated Multidimensional Spectrometry, is a Python application programming interface and command-line tool"
- [other] m/z 118.086255–1521.971475 with known CCS values: "m/z 118.086255–1521.971475 with known CCS values"
- [results] Conversion to mzML from several other formats can be performed using the free and open-source ProteoWizard msconvert utility: "Conversion to mzML from several other formats can be performed using the free and open-source ProteoWizard msconvert utility"