Skip to content

Latest commit

 

History

History
98 lines (71 loc) · 2.89 KB

usage-guide.md

File metadata and controls

98 lines (71 loc) · 2.89 KB

MS-DIAL Preprocessing - Usage Guide

Overview

MS-DIAL is an open-source software for mass spectrometry data processing. It provides a user-friendly GUI alternative to XCMS with built-in annotation capabilities and excellent lipidomics support.

Prerequisites

# Download MS-DIAL
# https://systemsomicslab.github.io/compms/msdial/main.html

# For data conversion
# ProteoWizard: https://proteowizard.sourceforge.io/

# Python (for exported data)
pip install pandas numpy

Quick Start

Tell your AI agent what you want to do:

  • "Process my LC-MS data with MS-DIAL and export the feature table"
  • "Load MS-DIAL alignment results for downstream analysis"

Example Prompts

Data Preparation

"Convert my vendor raw files to mzML format for MS-DIAL" "Help me set up MS-DIAL parameters for positive mode Orbitrap data"

Processing

"What peak detection parameters should I use for 10 ppm mass accuracy?" "Set up alignment using my QC sample as reference" "Configure gap filling to recover missing peak values"

Annotation

"Set up LipidBlast library for lipid annotation in MS-DIAL" "Configure MS/MS matching against MassBank spectra"

Export and Analysis

"Load my MS-DIAL alignment_result.csv into Python for analysis" "Parse MS-DIAL output and create a clean feature matrix" "Merge MS-DIAL peak table with sample metadata"

What the Agent Will Do

  1. Guide MS-DIAL parameter selection
  2. Help interpret processing results
  3. Load exported alignment results
  4. Parse feature annotations
  5. Create analysis-ready data matrices
  6. Integrate with downstream statistical analysis

Tips

  • Convert vendor formats to mzML using ProteoWizard before MS-DIAL
  • Use a QC sample as alignment reference for best results
  • Lower minimum peak height if few features detected
  • Check centroid vs profile mode matches your data type
  • LipidBlast library is excellent for lipidomics annotation

MS-DIAL vs XCMS

Feature MS-DIAL XCMS
Interface GUI + Console R package
Annotation Built-in Separate
Lipidomics Excellent Manual
Learning curve Lower Higher
Batch processing Console mode R scripts

Key Parameters

Parameter Typical Value Description
Minimum peak height 1000-10000 Intensity threshold
Mass accuracy 5-20 ppm Instrument dependent
RT tolerance 0.1-0.5 min Alignment window

Output Files

  • alignment_result.csv - Main feature table with annotations
  • peak_area_matrix.csv - Peak areas only
  • msp_output.msp - MS/MS spectra for library

Spectral Libraries

  • LipidBlast - In-silico lipid library (built-in)
  • MassBank - Experimental spectra
  • GNPS - Community library
  • HMDB - Human metabolome

References

  • MS-DIAL: doi:10.1038/nmeth.4512
  • MS-DIAL 4: doi:10.1038/s41592-023-01888-3
  • LipidBlast: doi:10.1038/nmeth.2442