A general-purpose baseline estimation algorithm for spectroscopic data.

Area of Science:

• Proteomics
• Analytical Chemistry
• Spectroscopy

Background:

• Modern proteomics technologies like nuclear magnetic resonance (NMR) and mass spectrometry generate large datasets.
• Extracting meaningful signal from background noise is a significant challenge in data analysis.
• Accurate baseline estimation is critical for effective signal extraction in spectroscopic data.

Purpose of the Study:

• To propose a novel algorithm, the BXR algorithm, for robust baseline estimation.
• To demonstrate the algorithm's adaptability to various spectroscopic data and technologies.
• To specifically tailor and evaluate the algorithm for matrix-assisted laser desorption/ionization Fourier transform ion cyclotron resonance mass spectrometry (FTICR-MS).

Main Methods:

• Development of the BXR algorithm for baseline estimation.
• Adaptation of the BXR algorithm for FTICR-MS data.
• Comparative performance analysis against existing baseline estimation algorithms.

Main Results:

• The BXR algorithm provides computationally efficient and robust baseline estimation.
• It effectively handles one-sided signal types common in NMR and mass spectrometry.
• The algorithm demonstrates improved performance compared to existing methods.

Conclusions:

• The BXR algorithm is a superior method for baseline estimation in proteomics data.
• Its adaptability makes it suitable for diverse spectroscopic applications.
• The algorithm is implemented in the R package FTICRMS for accessibility.