Ion mobility spectrometry with pre-separation by multi-capillary columns (MCC/IMS) has become an established inexpensive, non-invasive bioanalytics technology for detecting volatile organic compounds (VOCs) with various metabolomics applications in medical research. To pave the way for this technology towards daily usage in medical practice, different steps still have to be taken. With respect to modern biomarker research, one of the most important tasks is the automatic classification of patient-specific data sets into different groups, healthy or not, for instance. Although sophisticated machine learning methods exist, an inevitable preprocessing step is reliable and robust peak detection without manual intervention. In this work we evaluate four state-of-the-art approaches for automated IMS-based peak detection: local maxima search, watershed transformation with IPHEx, region-merging with VisualNow, and peak model estimation (PME).We manually generated Metabolites 2013, 3 278 a gold standard with the aid of a domain expert (manual) and compare the performance of the four peak calling methods with respect to two distinct criteria. We first utilize established machine learning methods and systematically study their classification performance based on the four peak detectors’ results. Second, we investigate the classification variance and robustness regarding perturbation and overfitting. Our main finding is that the power of the classification accuracy is almost equally good for all methods, the manually created gold standard as well as the four automatic peak finding methods. In addition, we note that all tools, manual and automatic, are similarly robust against perturbations. However, the classification performance is more robust against overfitting when using the PME as peak calling preprocessor. In summary, we conclude that all methods, though small differences exist, are largely reliable and enable a wide spectrum of real-world biomedical applications.
References
[1]
Westhoff, M.; Litterst, P.; Maddula, S.; B？deker, B.; Baumbach, J.I. Statistical and bioinformatical methods to differentiate chronic obstructive pulmonary disease (COPD) including lung cancer from healthy control by breath analysis using ion mobility spectrometry. Int. J. Ion Mobil. Spectrom. 2011, 14, 1–11, doi:10.1007/s12127-010-0055-4.
[2]
Baumbach, J.I.; Westhoff, M. Ion mobility spectrometry to detect lung cancer and airway infections. Spectrosc. Eur. 2006, 18, 22–27.
[3]
Perl, T.; Juenger, M.; Vautz, W.; Nolte, J.; Kuhns, M.; Zepelin, B.; Quintel, M. Detection of characteristic metabolites of Aspergillus fumigatus and Candida species using ion mobility spectrometry-metabolic profiling by volatile organic compounds. Mycoses 2011, 54, 828–837, doi:10.1111/j.1439-0507.2011.02037.x.
[4]
Ruzsanyi, V.; Mochalski, P.; Schmid, A.; Wiesenhofer, H.; Klieber, M.; Hinterhuber, H.; Amann, A. Ion mobility spectrometry for detection of skin volatiles. J. Chromatogr. B 2012, 911, 84–92, doi:10.1016/j.jchromb.2012.10.028.
[5]
Ruzsanyi, V.; Baumbach, J.I.; Sielemann, S.; Litterst, P.; Westhoff, M.; Freitag, L. Detection of human metabolites using multi-capillary columns coupled to ion mobility spectrometers. J. Chromatogr. A 2005, 1084, 145–151, doi:10.1016/j.chroma.2005.01.055.
[6]
Baumbach, J.I. Ion mobility spectrometry coupled with multi-capillary columns for metabolic profiling of human breath. J. Breath Res. 2009, 3, 1–16.
[7]
B & S Analytik GmbH. Available online: http://www.bs-analytik.de/ (accessed on 15 March 2013).
[8]
Purkhart, R.; Hillmann, A.; Graupner, R.; Becher, G. Detection of characteristic clusters in IMS-Spectrograms of exhaled air polluted with environmental contaminants. Int. J. Ion Mobil. Spectrom. 2012, 15, 1–6, doi:10.1007/s12127-011-0086-5.
[9]
B？deker, B.; Vautz, W.; Baumbach, J.I. Peak finding and referencing in MCC/IMS-data. Int. J. Ion Mobil. Spectrom. 2008, 11, 83–87, doi:10.1007/s12127-008-0012-7.
[10]
Bunkowski, A. MCC-IMS data analysis using automated spectra processing and explorative visualisation methods. PhD thesis, University Bielefeld, Bielefeld, Germany, 2011.
[11]
Kopczynski, D.; Baumbach, J.I.; Rahmann, S. Peak Modeling for Ion Mobility Spectrometry Measurements. In Proceedings of 20th European Signal Processing Conference, Bucharest, Romania, 27–31 August 2012.
[12]
Vogtland, D.; Baumbach, J.I. Breit-Wigner-function and IMS-signals. Int. J. Ion Mobil. Spectrom. 2009, 12, 109–114, doi:10.1007/s12127-009-0027-8.
[13]
Bader, S. Identification and Quantification of Peaks in Spectrometric Data. PhD thesis, TU Dortmund, Dortmund, Germany, 2008.
[14]
Nixon, M.; Aguado, A.S. Feature Extraction & Image Processing, 2nd ed. ed.; Academic Press: Waltham, MA, USA, 2008.
[15]
Savitzky, A.; Golay, M. Smoothing and differentiation of data by simplified least squares procedures. Anal. Chem. 1964, 36, 1627–1639, doi:10.1021/ac60214a047.
[16]
Hastie, T.; Tibshirani, R.; Friedman, J. The Elements of Statistical Learning: Data Mining, Inference, and Prediction; Springer: Berlin/Heidelberg, Germany, 2009.
[17]
Bader, S.; Urfer, W.; Baumbach, J.I. Reduction of ion mobility spectrometry data by clustering characteristic peak structures. J. Chemom. 2007, 20, 128–135, doi:10.1002/cem.998.
[18]
Vincent, L.; Soille, P. Watersheds in digital spaces: An efficient algorithm based on immersion simulations. IEEE Trans. Pattern Anal. Mach. Intell. 1991, 13, 583–598, doi:10.1109/34.87344.
[19]
Fong, S.S.; Rearden, P.; Kanchagar, C.; Sassetti, C.; Trevejo, J.; Brereton, R.G. Automated peak detection and matching algorithm for gas chromatography-differential mobility spectrometry. Anal. Chem. 2011, 83, 1537–1546, doi:10.1021/ac102110y.
[20]
Boser, B.; Guyon, I.; Vapnik, V. A training algorithm for optimal margin classifiers. In Proceedings of the Fifth Annual Workshop on Computational Learning Theory, Pittsburgh, USA, 27–29 July 1992; pp. 144–152.
[21]
Dimitriadou, E.; Hornik, K.; Leisch, F.; Meyer, D.; Weingessel, A. e1071: Misc Functions of the Department of Statistics (e1071), TU Wien; TU Wien: Vienna, Austria, 2010.
[22]
Liaw, A.; Wiener, M. Classification and regression by randomforest. R News 2002, 2, 18–22.
[23]
Robin, X.; Turck, N.; Hainard, A.; Tiberti, N.; Lisacek, F.; Sanchez, J.; M., M. pROC: An open-source package for R and S+ to analyze and compare ROC curves. BMC Bioinforma. 2011, 12, 77, doi:10.1186/1471-2105-12-77.
[24]
Ion Mobility Spectroscopy Analysis with Restricted Resources Home Page. Available online: http://www.rahmannlab.de/research/ims (accessed on 15 March 2013).
[25]
Collaborative Research Center SFB 876 -Providing Information by Resource-Constrained Data Analysis. Available online: http://sfb876.tu-dortmund.de (accessed on 15 March 2013).
