Clusterwise Peak Detection and Filtering Based on Spatial Distribution To Efficiently Mine Mass Spectrometry Imaging...

Area of Science:

• Biomedical Imaging
• Analytical Chemistry
• Computational Biology

Background:

• Mass spectrometry imaging (MSI) enables spatial mapping of biomolecules in tissues.
• Current MSI data processing methods, like binning, can lead to inaccurate signal extraction and low-quality ion images.
• High-resolution MSI instruments generate large datasets, necessitating advanced analytical techniques.

Purpose of the Study:

• To develop a novel data processing approach for MSI to improve the accuracy of biomolecule signal detection.
• To overcome limitations of traditional binning methods in MSI data analysis.
• To enhance the quality of ion images for clinically relevant information.

Main Methods:

• A novel data processing approach utilizing clusterwise kernel density estimates (KDEs) for peak detection.
• Sensitive peak detection to identify faint and localized signals.
• Comparison with existing binning-based methods for MSI data analysis.

Main Results:

• The proposed method successfully recalls more ground-truth molecules, fragments, and isotopes compared to binning methods.
• Accurate detection of faint and localized signals in MSI data.
• Automatic identification of previously reported lipid molecular ions, including those with close m/z values.

Conclusions:

• The novel KDE-based peak detection method significantly improves MSI data processing.
• This approach leads to higher quality ion images and more comprehensive biomolecule localization.
• The method offers a more accurate and sensitive alternative for analyzing complex MSI datasets.