Feature Extraction for Classification of Proteomic Profile

Abstract:

This work studies on feature extraction for classification of proteomic profile. We evaluated four methods, including principal component analysis (PCA), independent component analysis (ICA), locally linear embedding (LLE) and weighted maximum margin criterion (WMMC). PCA, ICA and LLE extract features based on traditional low-dimension map technique. Comparatively, WMMC extracts features according to classification goal. To study classification performance of PCA, ICA, LLE and WMMC in detail, we used two well known classification methods, support vector machine (SVM) and Fisher discriminant analysis (FDA), to classify profiles. The results show WMMC having relatively good performance due to its prediction accuracy, sensitivity and specificity for diagnosis; it can correctly identify features with high discrimination ability from high-dimensional proteomic profile. When feature set size was reduced less than 10, PCA, ICA and LLE lose a lot of classification information, and the prediction accuracies are less than 90%. However, WMMC can extract most classification information. Its prediction accuracies, sensitivities and specificities are more than 95%. Obviously, WMMC is more suitable to proteomic profile classification. For classifier, FDA is sensible to feature extraction.

• Periodical: Advanced Materials Research (Volumes 756-759)
• Pages: 4576-4580
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.756-759.4576
• Online since: September 2013
• Authors: Xiao Li Yang, Qiong He, Fen Yang
• Keywords: Feature Extraction, Independent Component Analysis (ICA), Locally Linear Embedding (LLE), Principal Component Analysis (PCA), Proteomic Profile, Weighted Maximum Margin Criterion (WMMC)

References

[1] Emanuel F Petricoin III, Ali M Ardekani, Ben A Hitt, et al. Use of proteomic patterns in serum to identify ovarian cancer [J]. Lancet 2002 359 572-577.

DOI: 10.1016/s0140-6736(02)07746-2

[2] Emanuel F Petricoin III and L. A. Liotta. SELDI-TOF-based serum proteomic pattern diagnosis for early detection of cancer [J]. Current Opinion in Biotechnology 2004 15(1) 24-30.

DOI: 10.1016/j.copbio.2004.01.005

[3] Julia D Wulfkuhle, Lance A Liotta and Emanuel F Petricoin. Proteomic applications for the early detection of cancer [J]. Nature Reviews Cancer 2003 3 267-275.

DOI: 10.1038/nrc1043

[4] L Lancashire, O Schmid, H Shah, and G Ball. Classification of bacterial species from proteomic data using combinatorial approaches incorporating artificial neural networks, cluster analysis and principal components analysis [J]. Bioinformatics 2005 21(102005) 2191-2199.

DOI: 10.1093/bioinformatics/bti368

[5] Oliver P Whelehan, Mark E Earll, Erik Johansson, et al. Detection of ovarian cancer using chemometric analysis of proteomic profiles [J]. Chemometrics and Intelligent Laboratory Systems 2006 84(1-2) 82-87.

DOI: 10.1016/j.chemolab.2006.03.008

[6] Yihui Liu and Li Bai. Ovarian cancer mass spectrometry data analysis based on ICA algorithm [J]. Future BioMedical Information Engineering 2008 (FBIE ' 08) 30-33.

DOI: 10.1109/fbie.2008.101

[7] Dante Mantini, Francesca Petrucci, Piero Del Boccio, et al. Independent component analysis for the extraction of reliable protein signal profiles from MALDI-TOF mass spectra [J]. Bioinformatics 2007 24 63-70.

DOI: 10.1093/bioinformatics/btm533

[8] Sam T Roweis and Lawrence K Saul. Nonlinear dimensionality reduction by locally linear embedding [J]. Science 2000 290(5500) 2323-2326.

DOI: 10.1126/science.290.5500.2323

[9] Xiaoli Yang, Qiong He, Siya Yang, et al. Weighted maximum margin criterion method: application to proteomic peptide profile [J]. 5th International Conference on Bioinformatics and Biomedical Engineering (iCBBE 2011) 2011 2 15-18.

DOI: 10.1109/icbbe.2011.5779976

[10] http: /home. ccr. cancer. gov/ncifdaproteomics/ppatterns. asp.

[11] Xiaoli Yang and Yanchun Yang. A novel baseline correction method for MALDI-TOF mass spectral [J]. Computers and Applied Chemistry 2010 27(11) 1498-1502.

[12] Deukwoo Kwon, Marina Vannucci, Joon Jin Song, et al. A novel wavelet-based thresholding method for the pre-processing of mass spectrometry data that accounts for heterogeneous noise [J] Proteomics 2008 8(15) 3019-3029.

DOI: 10.1002/pmic.200701010

[13] Jung Hun Oh, Jean Gao, Animesh Nandi, et al. Diagnosis of early relapse in ovarian cancer using serum proteomic profiling [J]. Genome Informatics 2002 16(2) 195-204.

[14] Kailin Tang, Tonghua Li, Wenwei Xiong, et al. Ovarian cancer classificatin based on dimensionality reduction for SELDI-TOF data [J]. BMC Bioinformatics 2010 11 109.

DOI: 10.1186/1471-2105-11-109

[15] Qingsong Xu and Yizeng Liang. Monte Carlo cross validation [J]. Chemometrics and Intelligent Laboratory Systems 2001 56(1) 1-11.

DOI: 10.1016/s0169-7439(00)00122-2