Ovarian cancer classification based on dimensionality reduction for SELDI-TOF data

Abstract

BackgroundRecent advances in proteomics technologies such as SELDI-TOF mass spectrometry has shown promise in the detection of early stage cancers. However, dimensionality reduction and classification are considerable challenges in statistical machine learning. We therefore propose a novel approach for dimensionality reduction and tested it using published high-resolution SELDI-TOF data for ovarian cancer.ResultsWe propose a method based on statistical moments to reduce feature dimensions. After refining and t-testing, SELDI-TOF data are divided into several intervals. Four statistical moments (mean, variance, skewness and kurtosis) are calculated for each interval and are used as representative variables. The high dimensionality of the data can thus be rapidly reduced. To improve efficiency and classification performance, the data are further used in kernel PLS models. The method achieved average sensitivity of 0.9950, specificity of 0.9916, accuracy of 0.9935 and a correlation coefficient of 0.9869 for 100 five-fold cross validations. Furthermore, only one control was misclassified in leave-one-out cross validation.ConclusionThe proposed method is suitable for analyzing high-throughput proteomics data.

Prostate Cancer Classification from Mass Spectrometry Data by Using Wavelet Analysis and Kernel Partial Least Squares Algorithm

“…In [8], it is claimed that statistical moments preserve the data properties while reducing the dimensionality. Although they achieved a reasonable classification performance, it is shown that statistical moments lose some of the useful discriminatory information when compared to the wavelet analysis.…”

“…In [8], the same statistical moments are used to classify ovarian cancer MS data. In the classification phase of this study, KPLS (Kernel Partial Least Square) method is used.…”

“…Recently, wavelet analysis and statistical moments are used for this purpose [6]- [8]. In the former one, the discrete wavelet transform (DWT) is applied to the MS data and approximation coefficients are obtained.…”

“…In the last stage, a feature transformation method, kernel partial least square (KPLS) is used. In [8], different from this study, KPLS is used as a classification stage after the dimensionality reduction stages for ovarian cancer identification.…”

Combining proteomics, serum biomarkers and bioinformatics to discriminate between esophageal squamous cell carcinoma and pre-cancerous lesion

“…Surfaceenhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF-MS) combined with ProteinChip technology is a high-throughput proteomic analysis approach which has been shown to be effective in the detection of biomarkers for early stage malignancy (Petricoin and Liotta, 2004). Using this combined technology, some new serum biomarkers with higher sensitivity were found for the early detection of different cancers, including prostate cancer (Okamoto et al, 2009;Yamamoto-Ishikawa et al, 2009), ovarian cancer (Høgdall et al, 2010;Tang et al, 2010), brain cancer , colorectal cancer (Yu et al, 2004;Helgason et al, 2010), breast cancer (Hu et al, 2005;Opstal-van Winden et al, 2011), lung cancer (Rathinam et al, 2011), and pancreatic cancer (Felix et al, 2011). However, reports on the protein profiling of different stages of ESCC development have not yet been published.…”