A novel preprocessing method using Hilbert Huang Transform for MALDI-TOF and SELDI-TOF mass spectrometry data

Area of Science:

• Biotechnology
• Analytical Chemistry
• Signal Processing

Background:

• Mass spectrometry is a crucial high-throughput technique for protein analysis, enabling comparisons between normal and disease groups.
• Mass spectrometry data is often complicated by scale shifting, non-stationarity, and significant noise, necessitating robust preprocessing.
• Effective preprocessing is vital for accurate analysis and reliable peak detection in mass spectrometry spectra.

Purpose of the Study:

• To develop and evaluate a novel preprocessing algorithm for MALDI and SELDI mass spectrometry spectra.
• To address challenges posed by noise and spectral shifting in mass spectrometry data analysis.
• To enhance the accuracy of protein peak detection and comparison between biological groups.

Main Methods:

• Utilized Hilbert-Huang Transformation (HHT), a non-stationary signal processing technique.
• Applied HHT to decompose spectra, filtering out high and low-frequency noise.
• Assessed the efficacy of HHT compared to wavelet and traditional preprocessing methods.

Main Results:

• HHT demonstrated superior performance in denoising mass spectrometry spectra compared to other methods.
• The HHT approach successfully identified protein peaks, even in complex datasets.
• While HHT processing is time-intensive, the resulting high-quality data justifies the duration.

Conclusions:

• Hilbert-Huang Transformation offers an efficient and effective method for denoising mass spectrometry data.
• HHT preprocessing significantly improves the reliability of protein identification and comparative analysis.
• The benefits of HHT in enhancing data quality outweigh its computational cost for complex spectra.