Quality assessment of tandem mass spectra using support vector machine (SVM)
2009

Assessing the Quality of Tandem Mass Spectra Using Support Vector Machine

Sample size: 37043 publication 10 minutes Evidence: high

Author Information

Author(s): Zou An-Min, Wu Fang-Xiang, Ding Jia-Rui, Poirier Guy G

Primary Institution: University of Saskatchewan

Hypothesis

Can a support vector machine (SVM) effectively assess the quality of tandem mass spectra?

Conclusion

The proposed SVM method effectively removes poor quality spectra before searching, improving peptide identification.

Supporting Evidence

  • The SVM classifiers can eliminate about 90% of poor quality spectra while losing less than 8% of high quality spectra.
  • The method outperformed existing methods in filtering poor quality spectra.
  • Training was conducted on two different datasets, ISB and TOV, demonstrating robustness.

Takeaway

This study shows how a computer program can help scientists find good quality data from messy results in mass spectrometry.

Methodology

The study used support vector machines to classify tandem mass spectra based on 16 features.

Potential Biases

The classifiers were trained on imbalanced datasets, which could affect performance.

Limitations

The method may still misclassify some spectra due to inherent noise in mass spectrometry data.

Digital Object Identifier (DOI)

10.1186/1471-2105-10-S1-S49

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication