A robust meta-classification strategy for cancer detection from MS data

We propose a novel method for phenotype identification involving a stringent noise analysis and filtering procedure followed by combining the results of several machine learning tools to produce a robust predictor. We illustrate our method on SELDI-TOF MS prostate cancer data (http://home.ccr.cancer.gov/ncifdaproteomics/ppatterns.asp). Our method identified 11 proteomic biomarkers and gave significantly improved predictions over previous analyses with these data. We were able to distinguish cancer from non-cancer cases with a sensitivity of 90.31% and a specificity of 98.81%. The proposed method can be generalized to multi-phenotype prediction and other types of data (e.g., microarray data).     

Serum protein profile in systemic-onset juvenile idiopathic arthritis differentiates response versus nonresponse to therapy

Systemic-onset juvenile idiopathic arthritis (SJIA) is a disease of unknown etiology with an unpredictable response to treatment. We examined two groups of patients to determine whether there are serum protein profiles reflective of active disease and predictive of response to therapy. The first group (n = 8) responded to conventional therapy. The second group (n = 15) responded to an experimental antibody to the IL-6 receptor (MRA). Paired sera from each patient were analyzed before and after treatment, using surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS). Despite the small number of patients, highly significant and consistent differences were observed before and after response to therapy in all patients. Of 282 spectral peaks identified, 23 had mean signal intensities significantly different (P < 0.001) before treatment and after response to treatment. The majority of these differences were observed regardless of whether patients responded to conventional therapy or to MRA. These peaks represent potential biomarkers of active disease. One such peak was identified as serum amyloid A, a known acute-phase reactant in SJIA, validating the SELDI-TOF MS platform as a useful technology in this context. Finally, profiles from serum samples obtained at the time of active disease were compared between the two patient groups. Nine peaks had mean signal intensities significantly different (P < 0.001) between active disease in patients who responded to conventional therapy and in patients who failed to respond, suggesting a possible profile predictive of response. Collectively, these data demonstrate the presence of serum proteomic profiles in SJIA that are reflective of active disease and suggest the feasibility of using the SELDI-TOF MS platform used as a tool for proteomic profiling and discovery of novel biomarkers in autoimmune diseases.     

The nucleotide-independent Fe(III)-binding site is located on beta subunit of the mitochondrial F(1)-ATPase

The need for methods to identify disease biomarkers is underscored by the survival-rate of patients diagnosed at early stages of cancer progression. Surface enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS) is a novel approach to biomarker discovery that combines two powerful techniques: chromatography and mass spectrometry. One of the key features of SELDI-TOF MS is its ability to provide a rapid protein expression profile from a variety of biological and clinical samples. It has been used for biomarker identification as well as the study of protein-protein, and protein-DNA interaction. The versatility of SELDI-TOF MS has allowed its use in projects ranging from the identification of potential diagnostic markers for prostate, bladder, breast, and ovarian cancers and Alzheimer's disease, to the study of biomolecular interactions and the characterization of posttranslational modifications. In this minireview we discuss the application of SELDI-TOF MS to protein biomarker discovery and profiling.     

Dependence network modeling for biomarker identification

MOTIVATION: Our purpose is to develop a statistical modeling approach for cancer biomarker discovery and provide new insights into early cancer detection. We propose the concept of dependence network, apply it for identifying cancer biomarkers, and study the difference between the protein or gene samples from cancer and non-cancer subjects based on mass-spectrometry (MS) and microarray data. RESULTS: Three MS and two gene microarray datasets are studied. Clear differences are observed in the dependence networks for cancer and non-cancer samples. Protein/gene features are examined three at one time through an exhaustive search. Dependence networks are constructed by binding triples identified by the eigenvalue pattern of the dependence model, and are further compared to identify cancer biomarkers. Such dependence-network-based biomarkers show much greater consistency under 10-fold cross-validation than the classification-performance-based biomarkers. Furthermore, the biological relevance of the dependence-network-based biomarkers using microarray data is discussed. The proposed scheme is shown promising for cancer diagnosis and prediction. AVAILABILITY: See supplements: http://dsplab.eng.umd.edu/~genomics/dependencenetwork/     

Proteomic detection of cancer in asbestosis patients using SELDI-TOF discovered serum protein biomarkers

Objectives: To identify biomarkers for cancer in asbestosis patients.

Methods: SELDI-TOF and CART were used to identify serum biomarker profiles in 35 asbestosis patients who subsequently developed cancer and 35 did not develop cancer.

Results: Three polypeptide peaks (5707.01, 6598.10, and 20,780.70?Da) could predict the development of cancer with 87% sensitivity and 70% specificity. The first two peaks were identified as KIF18A and KIF5A, respectively, and are part of the Kinesin Superfamily of proteins.

Conclusions: We identified two Kinesin proteins that can be potentially used as blood biomarkers to identify asbestosis patients at risk of developing lung cancer.     

Reproducibility of mass spectrometry based protein profiles for diagnosis of ovarian cancer across clinical studies: A systematic review

The focus of this systematic review is to give an overview of the current status of clinical protein profiling studies using MALDI and SELDI MS platforms in the search for ovarian cancer biomarkers. A total of 34 profiling studies were qualified for inclusion in the review. Comparative analysis of published discriminatory peaks to peaks found in an original MALDI MS protein profiling study was made to address the key question of reproducibility across studies. An overlap was found despite substantial heterogeneity between studies relating to study design, biological material, pre-analytical treatment, and data analysis. About 47% of the peaks reported to be associated to ovarian cancer were also represented in our experimental study, and 34% of these redetected peaks also showed a significant difference between cases and controls in our study. Thus, despite known problems related to reproducibility an overlap in peaks between clinical studies was demonstrated, which indicate convergence toward a set of common discriminating, reproducible peaks for ovarian cancer. The potential of the discriminating protein peaks for clinical use as ovarian cancer biomarkers will be discussed and evaluated. This article is part of a Special Issue entitled: Proteomics: The clinical link.     

Evaluation of proteomics-identified CCL18 and CXCL1 as circulating tumor markers for differential diagnosis between ovarian carcinomas and benign pelvic masses

A lack of sensitive and specific tumor markers for early diagnosis and treatment is a major cause for the high mortality rate of ovarian cancer. The purpose of this study was to identify potential proteomics-based biomarkers useful for the differential diagnosis between ovarian cancer and benign pelvic masses. Serum samples from 41 patients with ovarian cancer, 32 patients with benign pelvic masses, and 41 healthy female blood donors were examined, and proteomic profiling of the samples was assessed by surface-enhanced laser desorption/ionization time-of-flight (SELDI-TOF) mass spectroscopy (MS). A confirmatory study was also conducted with serum specimens from 58 patients with ovarian carcinoma, 37 patients with benign pelvic masses, and 48 healthy women. A classification tree was established using Biomarker Pattern Software. Six differentially expressed proteins (APP, CA 125, CCL18, CXCL1, IL-8, and ITIH4) were separated by high-performance liquid chromatography and identified by matrix-assisted laser desorption/ionization (MALDI)-MS/MS and database searches. Two of the proteins overexpressed in ovarian cancer patients, chemokine CC2 motif ligand 18 (CCL18) and chemokine CXC motif ligand 1 (CXCL1), were automatically selected in a multivariate predictive model. These two protein biomarkers were then validated and evaluated by enzyme-linked immunosorbent assay (ELISA) in 535 serum specimens (130 ovarian cancer, 64 benign ovarian masses, 36 lung cancer, 60 gastric cancer, 55 nasopharyngeal carcinoma, 48 hepatocellular carcinoma, and 142 healthy women). The combined use of CCL18 and CXCL1 as biomarkers for ovarian cancer had a sensitivity of 92% and a specificity of 97%. The multivariate ELISA analysis of the two putative markers in combination with CA 125 resulted in a sensitivity of 99% for healthy women and 94% for benign pelvic masses, and a specificity of 92% for both groups; these values were significantly higher than those obtained with CA 125 alone (p and lt;0.05). We conclude that serum CCL18 and CXCL1 are potentially useful as novel circulating tumor markers for the differential diagnosis between ovarian cancer and benign ovarian masses.     

Assessing the utility of SELDI-TOF and model averaging for serum proteomic biomarker discovery

The SELDI-TOF technique was used to profile serum proteins from Type 1 diabetes (T1D) patients and healthy autoantibody-negative (AbN) controls. Univariate and multivariate analyses were performed to identify putative biomarkers for T1D and to assess the reproducibility of the SELDI technique. We found 146 protein/peptide peaks (581 total peaks discovered) in human serum showing statistical differences in expression levels between T1D patients and controls, with 84% of these peaks showing technical replication. Because individual proteins did not offer great power for disease prediction, we used our model averaging approach that combines the information from multiple multivariate models to accurately classify T1D and control subjects (88.9% specificity and 90.0% sensitivity). Analyses of a test subset of the data showed less accuracy (82.8% specificity and 76.2% sensitivity), although the results are still positive. Unfortunately, no multivariate model could be replicated using the same samples. This first attempt of high throughput analyses of the human serum proteome in T1D patients suggests that model averaging is a viable method for developing biomarkers; however, the reproducibility of SELDI-TOF is currently not sufficient to be used for classification of complex diseases like T1D.     

Improving feature detection and analysis of surface-enhanced laser desorption/ionization-time of flight mass spectra

Discovering valid biological information from surface-enhanced laser desorption/ionization-time of flight mass spectrometry (SELDI-TOF MS) depends on clear experimental design, meticulous sample handling, and sophisticated data processing. Most published literature deals with the biological aspects of these experiments, or with computer-learning algorithms to locate sets of classifying biomarkers. The process of locating and measuring proteins across spectra has received less attention. This process should be tunable between sensitivity and false-discovery, and should guarantee that features are biologically meaningful in that they represent chemical species that can be identified and investigated. Existing feature detection in SELDI-TOF MS is not optimal for acquiring biologically relevant data. Most methods have so many user-defined settings that reproducibility and comparability among studies suffer considerably. To address these issues, we have developed an approach, called simultaneous spectrum analysis (SSA), which (i) locates proteins across spectra, (ii) measures their abundance, (iii) subtracts baseline, (iv) excludes irreproducible measurements, and (v) computes normalization factors for comparing spectra. SSA uses only two key parameters for feature detection and one parameter each for quality thresholds on spectra and peaks. The effectiveness of SSA is demonstrated by identifying proteins differentially expressed in SELDI-TOF spectra from plasma of wild-type and knockout mice for plasma glutathione peroxidase. Comparing analyses by SSA and CiphergenExpress Data Manager 2.1 finds similar results for large signal peaks, but SSA improves the number and quality of differences betweens groups among lower signal peaks. SSA is also less likely to introduce systematic bias when normalizing spectra.     

Proteomic biomarker identification for diagnosis of early relapse in ovarian cancer

Ovarian cancer recurs at the rate of 75% within a few months or several years later after therapy. Early recurrence, though responding better to treatment, is difficult to detect. Surface-enhanced laser desorption/ionization time-of-flight (SELDI-TOF) mass spectrometry has showed the potential to accurately identify disease biomarkers to help early diagnosis. A major challenge in the interpretation of SELDI-TOF data is the high dimensionality of the feature space. To tackle this problem, we have developed a multi-step data processing method composed of t-test, binning and backward feature selection. A new algorithm, support vector machine-Markov blanket/recursive feature elimination (SVM-MB/RFE) is presented for the backward feature selection. This method is an integration of minimum weight feature elimination by SVM-RFE and information theory based redundant/irrelevant feature removal by Markov Blanket. Subsequently, SVM was used for classification. We conducted the biomarker selection algorithm on 113 serum samples to identify early relapse from ovarian cancer patients after primary therapy. To validate the performance of the proposed algorithm, experiments were carried out in comparison with several other feature selection and classification algorithms.     

A robust biomarker discovery pipeline for high-performance mass spectrometry data

A high-throughput software pipeline for analyzing high-performance mass spectral data sets has been developed to facilitate rapid and accurate biomarker determination. The software exploits the mass precision and resolution of high-performance instrumentation, bypasses peak-finding steps, and instead uses discrete m/z data points to identify putative biomarkers. The technique is insensitive to peak shape, and works on overlapping and non-Gaussian peaks which can confound peak-finding algorithms. Methods are presented to assess data set quality and the suitability of groups of m/z values that map to peaks as potential biomarkers. The algorithm is demonstrated with serum mass spectra from patients with and without ovarian cancer. Biomarker candidates are identified and ranked by their ability to discriminate between cancer and noncancer conditions. Their discriminating power is tested by classifying unknowns using a simple distance calculation, and a sensitivity of 95.6% and a specificity of 97.1% are obtained. In contrast, the sensitivity of the ovarian cancer blood marker CA125 is approximately 50% for stage I/II and approximately 80% for stage III/IV cancers. While the generalizability of these markers is currently unknown, we have demonstrated the ability of our analytical package to extract biomarker candidates from high-performance mass spectral data.     

In-source decay causes artifacts in SELDI-TOF MS spectra

SELDI-TOF MS is a mass spectrometric technique which has been extensively used for biomarker discovery. In this study, we show that in-source decay is an important source for the generation of additional spectral peaks with this technique, both for pure proteins and proteins in serum samples. Thus, SELDI-TOF MS could be used to gain sequence information from proteins, but the results also question the uncritical use of SELDI-TOF MS as a general method for the detection of biomarkers.     

Support vector machine classification and validation of cancer tissue samples using microarray expression data

MOTIVATION: DNA microarray experiments generating thousands of gene expression measurements, are being used to gather information from tissue and cell samples regarding gene expression differences that will be useful in diagnosing disease. We have developed a new method to analyse this kind of data using support vector machines (SVMs). This analysis consists of both classification of the tissue samples, and an exploration of the data for mis-labeled or questionable tissue results. RESULTS: We demonstrate the method in detail on samples consisting of ovarian cancer tissues, normal ovarian tissues, and other normal tissues. The dataset consists of expression experiment results for 97,802 cDNAs for each tissue. As a result of computational analysis, a tissue sample is discovered and confirmed to be wrongly labeled. Upon correction of this mistake and the removal of an outlier, perfect classification of tissues is achieved, but not with high confidence. We identify and analyse a subset of genes from the ovarian dataset whose expression is highly differentiated between the types of tissues. To show robustness of the SVM method, two previously published datasets from other types of tissues or cells are analysed. The results are comparable to those previously obtained. We show that other machine learning methods also perform comparably to the SVM on many of those datasets. AVAILABILITY: The SVM software is available at http://www.cs. columbia.edu/ approximately bgrundy/svm.     

A proteome resource of ovarian cancer ascites: integrated proteomic and bioinformatic analyses to identify putative biomarkers

Epithelial ovarian cancer is the most lethal gynecological malignancy, and disease-specific biomarkers are urgently needed to improve diagnosis, prognosis, and to predict and monitor treatment efficiency. We present an in-depth proteomic analysis of selected biochemical fractions of human ovarian cancer ascites, resulting in the stringent and confident identification of over 2500 proteins. Rigorous filter schemes were applied to objectively minimize the number of false-positive identifications, and we only report proteins with substantial peptide evidence. Integrated computational analysis of the ascites proteome combined with several recently published proteomic data sets of human plasma, urine, 59 ovarian cancer related microarray data sets, and protein-protein interactions from the Interologous Interaction Database I (2)D ( http://ophid.utoronto.ca/i2d) resulted in a short-list of 80 putative biomarkers. The presented proteomics analysis provides a significant resource for ovarian cancer research, and a framework for biomarker discovery.     

Identification of treatment efficacy-related host factors in chronic hepatitis C by ProteinChip serum analysis

Recent development of proteomic array technology, including protein profiling coupling ProteinChip array with surface-enhanced laser desorption ionization time-of-flight mass spectrometry (SELDI-TOF/MS), provides a potentially powerful tool for discovery of new biomarkers by comparison of its profiles according to patient phenotypes. We used this approach to identify the host factors associated with treatment response in patients with chronic hepatitis C (CHC) receiving a 48-wk course of pegylated interferon (PEG-IFN) alpha 2b plus ribavirin (RBV). Protein profiles of pretreatment serum samples from 32 patients with genotype 1b and high viral load were conducted by SELDI-TOF/MS by using the three different ProteinChip arrays (CM10, Q10, IMAC30). Proteins showed significantly different peak intensities between sustained virological responders (SVRs), and non-SVRs were identified by chromatography, SDS-PAGE, TOF/MS and tandem mass spectrometry (MS/MS) assay. Eleven peak intensities were significantly different between SVRs and non-SVRs. The three SVR-increased peaks could be identified as two apolipoprotein (Apo) fragments and albumin and, among the eight non-SVR-increased proteins, four peaks identified as two iron-related and two fibrogenesis-related protein fragments, respectively. Multivariate analysis showed that the serum ferritin and three peak intensity values (Apo A1, hemopexin and transferrin) were independent variables associated with SVRs, and the area under the receiver operating characteristic (ROC) curves for SVR prediction by using the Apo A1/hemopexin and hemopexin/transferrin were 0.964 and 0.936. In conclusion, pretreatment serum protein profiling by SELDI-TOF/MS is variable for identification of response-related host factors, which are useful for treatment efficacy prediction in CHC receiving PEG-IFN plus RBV. Our data also may help us understand the mechanism for treatment resistance and development of more effective antiviral therapy targeted toward the modulation of lipogenesis or iron homeostasis in CHC patients.     

A computational platform for MALDI-TOF mass spectrometry data: Application to serum and plasma samples

BackgroundMass spectrometry (MS) is becoming the gold standard for biomarker discovery. Several MS-based bioinformatics methods have been proposed for this application, but the divergence of the findings by different research groups on the same MS data suggests that the definition of a reliable method has not been achieved yet. In this work, we propose an integrated software platform, MASCAP, intended for comparative biomarker detection from MALDI-TOF MS data.ResultsMASCAP integrates denoising and feature extraction algorithms, which have already shown to provide consistent peaks across mass spectra; furthermore, it relies on statistical analysis and graphical tools to compare the results between groups. The effectiveness in mass spectrum processing is demonstrated using MALDI-TOF data, as well as SELDI-TOF data. The usefulness in detecting potential protein biomarkers is shown comparing MALDI-TOF mass spectra collected from serum and plasma samples belonging to the same clinical population.ConclusionsThe analysis approach implemented in MASCAP may simplify biomarker detection, by assisting the recognition of proteomic expression signatures of the disease. A MATLAB implementation of the software and the data used for its validation are available at http://www.unich.it/proteomica/bioinf.     

Proteomic biomarkers predicting lymph node involvement in serum of cervical cancer patients. Limitations of SELDI-TOF MS

ABSTRACT: BACKGROUND: Lymph node status is not part of the staging system for cervical cancer, but provides important information for prognosis and treatment. We investigated whether lymph node status can be predicted with proteomic profiling. Material & methods Serum samples of 60 cervical cancer patients (FIGO I/II) were obtained before primary treatment. Samples were run through a HPLC depletion column, eliminating the 14 most abundant proteins ubiquitously present in serum. Unbound fractions were concentrated with spin filters. Fractions were spotted onto CM10 and IMAC30 surfaces and analyzed with surface-enhanced laser desorption time of flight (SELDI-TOF) mass spectrometry (MS). Unsupervised peak detection and peak clustering was performed using MASDA software. Leave-one-out (LOO) validation for weighted Least Squares Support Vector Machines (LSSVM) was used for prediction of lymph node involvement. Other outcomes were histological type, lymphvascular space involvement (LVSI) and recurrent disease. RESULTS: LSSVM models were able to determine LN status with a LOO area under the receiver operating characteristics curve (AUC) of 0.95, based on peaks with m/z values 2,698.9, 3,953.2, and 15,254.8. Furthermore, we were able to predict LVSI (AUC 0.81), to predict recurrence (AUC 0.92), and to differentiate between squamous carcinomas and adenocarcinomas (AUC 0.88), between squamous and adenosquamous carcinomas (AUC 0.85), and between adenocarcinomas and adenosquamous carcinomas (AUC 0.94). CONCLUSIONS: Potential markers related with lymph node involvement were detected, and protein/peptide profiling support differentiation between various subtypes of cervical cancer. However, identification of the potential biomarkers was hampered by the technical limitations of SELDI-TOF MS.