On the analysis of glycomics mass spectrometry data via the regularized area under the ROC curve

Background  

Novel molecular and statistical methods are in rising demand for disease diagnosis and prognosis with the help of recent advanced biotechnology. High-resolution mass spectrometry (MS) is one of those biotechnologies that are highly promising to improve health outcome. Previous literatures have identified some proteomics biomarkers that can distinguish healthy patients from cancer patients using MS data. In this paper, an MS study is demonstrated which uses glycomics to identify ovarian cancer. Glycomics is the study of glycans and glycoproteins. The glycans on the proteins may deviate between a cancer cell and a normal cell and may be visible in the blood. High-resolution MS has been applied to measure relative abundances of potential glycan biomarkers in human serum. Multiple potential glycan biomarkers are measured in MS spectra. With the objection of maximizing the empirical area under the ROC curve (AUC), an analysis method was considered which combines potential glycan biomarkers for the diagnosis of cancer.     

Simultaneous and exact interval estimates for the contrast of two groups based on an extremely high dimensional variable: application to mass spec data

MOTIVATION: Analysis of high-throughput proteomic/genomic data, in particular, surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS) data and microarray data, has led to a multitude of techniques aimed at identifying potential biomarkers. Most of the statistical techniques for comparing two groups are based on qualitative measures such as P-value. A quantitative way such as interval estimation for the contrasts of two groups is more appealing. RESULTS: We have devised a simultaneous confidence bands method capable of detecting potential biomarkers, while controlling for overall confidence coverage level, in high-dimensional datasets that discriminate two treatment groups using a permutation scheme. For example, for the SELDI-TOF MS data, we deal with the entire spectrum simultaneously and construct (1 - alpha) confidence bands for the mean differences between groups. Furthermore, peaks were identified based on the maximal differences between the groups as determined by the confidence bands. The analysis method herein described gives both qualitative (P-value) and quantitative data (magnitude of difference). The Clinical Proteomics Programs Databank's ovarian cancer dataset and data from in-house samples containing known spiked-in proteins were analyzed. We were able to identify potential biomarkers similar to those described in previous analysis of the ovarian cancer data, however, while these markers are highly significant between cancer and normal groups, our analysis indicated the absolute difference between the two groups was minimal. In addition, we found additional markers than those previously described with greater differences in average intensities. The proposed confidence bands method successfully detected the spiked-in peaks, as well as, secondary peaks generated by adducts and double-charged species. We also illustrate our method utilizing paired gene expression data from a prostate cancer microarray experiment by constructing confidence bands for the fold changes between cancer and normal samples. AVAILABILITY: R-package, 'seie.zip' (license: GNU GPL), is publiclly available at http://research2.dfci.harvard.edu/dfci/MS_spike-in_data/     

Investigation of ovarian cancer associated sialylation changes in N-linked glycopeptides by quantitative proteomics

ABSTRACT: BACKGROUND: In approximately 80% of patients, ovarian cancer is diagnosed when the patient is already in the advanced stages of the disease. CA125 is currently used as the marker for ovarian cancer; however, it lacks specificity and sensitivity for detecting early stage disease. There is a critical unmet need for sensitive and specific routine screening tests for early diagnosis that can reduce ovarian cancer lethality by reliably detecting the disease at its earliest and treatable stages. Results: In this study, we investigated the N-linked sialylated glycopeptides in serum samples from healthy and ovarian cancer patients using Lectin-directed Tandem Labeling (LTL) and iTRAQ quantitative proteomics methods. We identified 45 N-linked sialylated glycopeptides containing 46 glycosylation sites. Among those, ten sialylated glycopeptides were significantly up-regulated in ovarian cancer patients' serum samples. LC-MS/MS analysis of the non-glycosylated peptides from the same samples, western blot data using lectin enriched glycoproteins of various ovarian cancer type samples, and PNGase F (+/-) treatment confirmed the sialylation changes in the ovarian cancer samples. Conclusion: Herein, we demonstrated that several proteins are aberrantly sialylated in N-linked glycopeptides in ovarian cancer and detection of glycopeptides with abnormal sialylation changes may have the potential to serve as biomarkers for ovarian cancer.     

Decon2LS: An open-source software package for automated processing and visualization of high resolution mass spectrometry data

Background

The majority of ovarian cancer biomarker discovery efforts focus on the identification of proteins that can improve the predictive power of presently available diagnostic tests. We here show that metabolomics, the study of metabolic changes in biological systems, can also provide characteristic small molecule fingerprints related to this disease.

Results

In this work, new approaches to automatic classification of metabolomic data produced from sera of ovarian cancer patients and benign controls are investigated. The performance of support vector machines (SVM) for the classification of liquid chromatography/time-of-flight mass spectrometry (LC/TOF MS) metabolomic data focusing on recognizing combinations or "panels" of potential metabolic diagnostic biomarkers was evaluated. Utilizing LC/TOF MS, sera from 37 ovarian cancer patients and 35 benign controls were studied. Optimum panels of spectral features observed in positive or/and negative ion mode electrospray (ESI) MS with the ability to distinguish between control and ovarian cancer samples were selected using state-of-the-art feature selection methods such as recursive feature elimination and L1-norm SVM.

Conclusion

Three evaluation processes (leave-one-out-cross-validation, 12-fold-cross-validation, 52-20-split-validation) were used to examine the SVM models based on the selected panels in terms of their ability for differentiating control vs. disease serum samples. The statistical significance for these feature selection results were comprehensively investigated. Classification of the serum sample test set was over 90% accurate indicating promise that the above approach may lead to the development of an accurate and reliable metabolomic-based approach for detecting ovarian cancer.     

Quantitative Detection of 15 Serum Bile Acid Metabolic Products by LC/MS/MS in the Diagnosis of Primary Biliary Cholangitis

To determine 15 bile acid metabolic products in human serum by liquid chromatography-tandem mass spectrometry (LC/MS/MS) and value their diagnostic outcome in primary biliary cholangitis (PBC). Serum from 20 healthy controls and 26 patients with PBC were collected and went LC/MS/MS analysis of 15 bile acid metabolic products. The test results were analyzed by bile acid metabolomics, and the potential biomarkers were screened and their diagnostic performance was judged by statistical methods such as principal component and partial least squares discriminant analysis and area under curve (AUC). 8 differential metabolites can be screened out: Deoxycholic acid (DCA), Glycine deoxycholic acid (GDCA), Lithocholic acid (LCA), Glycine ursodeoxycholic acid (GUDCA), Taurolithocholic acid (TLCA), Tauroursodeoxycholic acid (TUDCA), Taurodeoxycholic acid (TDCA), Glycine chenodeoxycholic acid (GCDCA). The performance of the biomarkers was evaluated by the AUC, specificity and sensitivity. In conclusion, DCA, GDCA, LCA, GUDCA, TLCA, TUDCA, TDCA and GCDCA were identified as eight potential biomarkers to distinguish between healthy people and PBC patients by multivariate statistical analysis, which provided reliable experimental basis for clinical practice.     

High-sensitivity blood-based detection of breast cancer by multi photon detection diagnostic proteomics

We have developed several new methods for blood-based cancer detection by diagnostic proteomics. Ultrasensitive methods of immunoassay using multiphoton-detection (IA/MPD) increase sensitivity by 200- to 1,000-fold (1 femtogram/mL). This has allowed the measurement of cancer biomarkers with very low concentrations in blood that could not be measured for full patient cohorts with conventional immunoassays. Sensitivity and specificity in cancer detection have been found to be potentiated by use of immunoassay panels which include tissue-specific cancer biomarkers as well as cytokines and angiogenic factors. The ultrasensitive immunoassays revealed that patient to patient variations in the concentrations of individual biomarkers in blood can extend over many orders of magnitude (up to six) and that the distributions of biomarker concentrations over patient cohorts are non-Gaussian. New methods of data analysis which correlate abundances of multiple, different biomarkers have been developed to deal with such data sets. Sensitivity and specificity of about 95% have been achieved for blood-based detection of breast cancer in pilot studies on 250 patients and 95 controls. Pilot studies indicate that this methodology may also allow differentiation of malignant breast cancer from benign lesions and can provide similar sensitivity and specificity for other epithelial cancers such as prostate cancer, ovarian cancer and melanoma. The methods developed for selection, application, and evaluation of very high sensitivity biomarker panels are expected to have general relevance for diagnostic proteomics.     

LC‐MS/MS‐based serum proteomics for identification of candidate biomarkers for hepatocellular carcinoma

Associating changes in protein levels with the onset of cancer has been widely investigated to identify clinically relevant diagnostic biomarkers. In the present study, we analyzed sera from 205 patients recruited in the United States and Egypt for biomarker discovery using label‐free proteomic analysis by LC‐MS/MS. We performed untargeted proteomic analysis of sera to identify candidate proteins with statistically significant differences between hepatocellular carcinoma (HCC) and patients with liver cirrhosis. We further evaluated the significance of 101 proteins in sera from the same 205 patients through targeted quantitation by MRM on a triple quadrupole mass spectrometer. This led to the identification of 21 candidate protein biomarkers that were significantly altered in both the United States and Egyptian cohorts. Among the 21 candidates, ten were previously reported as HCC‐associated proteins (eight exhibiting consistent trends with our observation), whereas 11 are new candidates discovered by this study. Pathway analysis based on the significant proteins reveals upregulation of the complement and coagulation cascades pathway and downregulation of the antigen processing and presentation pathway in HCC cases versus patients with liver cirrhosis. The results of this study demonstrate the power of combining untargeted and targeted quantitation methods for a comprehensive serum proteomic analysis, to evaluate changes in protein levels and discover novel diagnostic biomarkers. All MS data have been deposited in the ProteomeXchange with identifier PXD001171 ( http://proteomecentral.proteomexchange.org/dataset/PXD001171 ).     

Temporal characterization of serum metabolite signatures in lung cancer patients undergoing treatment

Lung cancer causes more deaths in men and women than any other cancer related disease. Currently, few effective strategies exist to predict how patients will respond to treatment. We evaluated the serum metabolomic profiles of 25 lung cancer patients undergoing chemotherapy ± radiation to evaluate the feasibility of metabolites as temporal biomarkers of clinical outcomes. Serial serum specimens collected prospectively from lung cancer patients were analyzed using both nuclear magnetic resonance (¹H-NMR) spectroscopy and gas chromatography mass spectrometry (GC–MS). Multivariate statistical analysis consisted of unsupervised principal component analysis or orthogonal partial least squares discriminant analysis with significance assessed using a cross-validated ANOVA. The metabolite profiles were reflective of the temporal distinction between patient samples before during and after receiving therapy (¹H-NMR, p < 0.001: and GC–MS p < 0.01). Disease progression and survival were strongly correlative with the GC–MS metabolite data whereas stage and cancer type were associated with ¹H-NMR data. Metabolites such as hydroxylamine, tridecan-1-ol, octadecan-1-ol, were indicative of survival (GC–MS p < 0.05) and metabolites such as tagatose, hydroxylamine, glucopyranose, and threonine that were reflective of progression (GC–MS p < 0.05). Metabolite profiles have the potential to act as prognostic markers of clinical outcomes for lung cancer patients. Serial ¹H-NMR measurements appear to detect metabolites diagnostic of tumor pathology, while GC–MS provided data better related to prognostic clinical outcomes, possibility due to physiochemical bias related to specific biochemical pathways. These results warrant further study in a larger cohort and with various treatment options.     

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/     

The knowledge-integrated network biomarkers discovery for major adverse cardiac events

The mass spectrometry (MS) technology in clinical proteomics is very promising for discovery of new biomarkers for diseases management. To overcome the obstacles of data noises in MS analysis, we proposed a new approach of knowledge-integrated biomarker discovery using data from Major Adverse Cardiac Events (MACE) patients. We first built up a cardiovascular-related network based on protein information coming from protein annotations in Uniprot, protein-protein interaction (PPI), and signal transduction database. Distinct from the previous machine learning methods in MS data processing, we then used statistical methods to discover biomarkers in cardiovascular-related network. Through the tradeoff between known protein information and data noises in mass spectrometry data, we finally could firmly identify those high-confident biomarkers. Most importantly, aided by protein-protein interaction network, that is, cardiovascular-related network, we proposed a new type of biomarkers, that is, network biomarkers, composed of a set of proteins and the interactions among them. The candidate network biomarkers can classify the two groups of patients more accurately than current single ones without consideration of biological molecular interaction.     

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.     

Identification of free fatty acids profiling of type 2 diabetes mellitus and exploring possible biomarkers by GC–MS coupled with chemometrics

Free fatty acids (FFAs), which are considered to be closely related with type 2 diabetes mellitus (T2DM), are not only the main energy source as nutrients, but also signaling molecules in insulin secretion. In this study, gas chromatography–mass spectrometry (GC–MS) coupled with two chemometric resolution methods, heuristic evolving latent projections (HELP) and selective ion analysis (SIA), was successfully applied to investigate plasma FFAs profiling of T2DM. Totally, twenty-three FFAs were identified and quantified. The results showed that HELP and SIA methods could be used to effectively handle overlapping peaks of GC–MS data and hence improve the qualitative and quantitative accuracy. Furthermore, a newly proposed competitive adaptive reweighted sampling (CARS) method coupled with partial least squares linear discriminant analysis (PLS-LDA) was introduced to seek the potential biomarkers. Finally, three fatty acids, oleic acid (OLA C18:1n-9), α-linolenic acid (ALA C18:3n-3), and eicosapentaenoic acid (EPA C20:5n-3), were identified as the potential biomarkers of T2DM for their powerful discriminant ability of T2DM patients from healthy controls. The study indicated that GC–MS combining with chemometric methods was a useful strategy to analyze metabolites and further screen the potential biomarkers of T2DM.     

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.     

The ovarian cancer-derived secretory/releasing proteome: A repertoire of tumor markers

Ovarian cancer is the most lethal gynecological malignancy worldwide, and early detection of this disease using serum or plasma biomarkers may improve its clinical outcome. In the present study, a large scale protein database derived from ovarian cancer was created to enable tumor marker discovery. First, primary organ cultures were established with the tumor tissues and corresponding normal tissues obtained from six ovarian cancer patients, and the serum-free conditioned medium (CM) samples were collected for proteomic analysis. The total proteins from the CM sample were separated by SDS-PAGE, digested with trypsin and then analyzed by LC-MS/MS. Combining data from the tumor tissues and the normal tissues, 1129 proteins were identified in total, of which those categorized as "extracellular proteins" and "plasma membrane proteins" accounted for 21.4% and 16.9%, respectively. For validation, three secretory proteins (NID1, TIMP2, and VCAN) involved in "organ development"-associated subnetwork, showed significant differences between their levels in the circulating plasma samples from ovarian cancer patients and healthy women. In conclusion, this ovarian cancer-derived protein database provides a credible repertoire of potential biomarkers in blood for this malignant disease, and deserves mining further.     

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.     

Pathway-based classification of cancer subtypes

ABSTRACT: BACKGROUND: Molecular markers based on gene expression profiles have been used in experimental and clinical settings to distinguish cancerous tumors in stage, grade, survival time, metastasis, and drug sensitivity. However, most significant gene markers are unstable (not reproducible) among data sets. We introduce a standardized method for representing cancer markers as 2-level hierarchical feature vectors, with a basic gene level as well as a second level of (more stable) pathway markers, for the purpose of discriminating cancer subtypes. This extends standard gene expression arrays with new pathway-level activation features obtained directly from off-the-shelf gene set enrichment algorithms such as GSEA. Such so-called pathway-based expression arrays are significantly more reproducible across datasets. Such reproducibility will be important for clinical usefulness of genomic markers, and augment currently accepted cancer classification protocols. RESULTS: The present method produced more stable (reproducible) pathway-based markers for discriminating breast cancer metastasis and ovarian cancer survival time. Between two datasets for breast cancer metastasis, the intersection of standard significant gene biomarkers totaled 7.47% of selected genes, compared to 17.65% using pathway-based markers; the corresponding percentages for ovarian cancer datasets were 20.65% and 33.33% respectively. Three pathways, consisting of Type_1_diabetes mellitus, Cytokine-cytokine_receptor_interaction and Hedgehog_signaling (all previously implicated in cancer), are enriched in both the ovarian long survival and breast non-metastasis groups. In addition, integrating pathway and gene information, we identified five (ID4, ANXA4, CXCL9, MYLK, FBXL7) and six (SQLE, E2F1, PTTG1, TSTA3, BUB1B, MAD2L1) known cancer genes significant for ovarian and breast cancer respectively. CONCLUSIONS: Standardizing the analysis of genomic data in the process of cancer staging, classification and analysis is important as it has implications for both pre-clinical as well as clinical studies. The paradigm of diagnosis and prediction using pathway-based biomarkers as features can be an important part of the process of biomarker-based cancer analysis, and the resulting canonical (clinically reproducible) biomarkers can be important in standardizing genomic data. We expect that identification of such canonical biomarkers will improve clinical utility of high-throughput datasets for diagnostic and prognostic applications. Reviewers This article was reviewed by John McDonald (nominated by I. King Jordon), Eugene Koonin, Nathan Bowen (nominated by I, King Jordon), and Ekaterina Kotelnikova (nominated by Mikhail Gelfand).     

Multiclass cancer classification and biomarker discovery using GA-based algorithms

MOTIVATION: The development of microarray-based high-throughput gene profiling has led to the hope that this technology could provide an efficient and accurate means of diagnosing and classifying tumors, as well as predicting prognoses and effective treatments. However, the large amount of data generated by microarrays requires effective reduction of discriminant gene features into reliable sets of tumor biomarkers for such multiclass tumor discrimination. The availability of reliable sets of biomarkers, especially serum biomarkers, should have a major impact on our understanding and treatment of cancer. RESULTS: We have combined genetic algorithm (GA) and all paired (AP) support vector machine (SVM) methods for multiclass cancer categorization. Predictive features can be automatically determined through iterative GA/SVM, leading to very compact sets of non-redundant cancer-relevant genes with the best classification performance reported to date. Interestingly, these different classifier sets harbor only modest overlapping gene features but have similar levels of accuracy in leave-one-out cross-validations (LOOCV). Further characterization of these optimal tumor discriminant features, including the use of nearest shrunken centroids (NSC), analysis of annotations and literature text mining, reveals previously unappreciated tumor subclasses and a series of genes that could be used as cancer biomarkers. With this approach, we believe that microarray-based multiclass molecular analysis can be an effective tool for cancer biomarker discovery and subsequent molecular cancer diagnosis.