Prediction of colorectal cancer diagnosis based on circulating plasma proteins

Diagnosis of major depressive disorder by combining multimodal information from heart rate dynamics and serum proteomics using machine-learning algorithm

OBJECTIVE

Major depressive disorder (MDD) is a systemic and multifactorial disorder that involves abnormalities in multiple biochemical pathways and the autonomic nervous system. This study applied a machine-learning method to classify MDD and control groups by incorporating data from serum proteomic analysis and heart rate variability (HRV) analysis for the identification of novel peripheral biomarkers.

METHODS

The study subjects consisted of 25 drug-free female MDD patients and 25 age- and sex-matched healthy controls. First, quantitative serum proteome profiles were analyzed by liquid chromatography-tandem mass spectrometry using pooled serum samples from 10 patients and 10 controls. Next, candidate proteins were quantified with multiple reaction monitoring (MRM) in 50 subjects. We also analyzed 22 linear and nonlinear HRV parameters in 50 subjects. Finally, we identified a combined biomarker panel consisting of proteins and HRV indexes using a support vector machine with recursive feature elimination.

RESULTS

A separation between MDD and control groups was achieved using five parameters (apolipoprotein B, group-specific component, ceruloplasmin, RMSSD, and SampEn) at 80.1% classification accuracy. A combination of HRV and proteomic data achieved better classification accuracy.

CONCLUSIONS

A high classification accuracy can be achieved by combining multimodal information from heart rate dynamics and serum proteomics in MDD. Our approach can be helpful for accurate clinical diagnosis of MDD. Further studies using larger, independent cohorts are needed to verify the role of these candidate biomarkers for MDD diagnosis.

Determination of differentially regulated proteins upon proteasome inhibition in AML cell lines by the combination of large-scale and targeted quantitative proteomics

The ubiquitin-proteasome pathway (UPP) plays a critical role in the degradation of proteins implicated in cell cycle control, signal transduction, DNA damage response, apoptosis and immune response. Proteasome inhibitors can inhibit the growth of a broad spectrum of human cancer cells by altering the balance of intracellular proteins. However, the targets of these compounds in acute myeloid leukemia (AML) cells have not been fully characterized. Herein, we combined large-scale quantitative analysis by SILAC-MS and targeted quantitative proteomic analysis in order to identify proteins regulated upon proteasome inhibition in two AML cell lines displaying different stages of maturation: immature KG1a cells and mature U937 cells. In-depth data analysis enabled accurate quantification of more than 7000 proteins in these two cell lines. Several candidates were validated by selected reaction monitoring (SRM) measurements in a large number of samples. Despite the broad range of proteins known to be affected by proteasome inhibition, such as heat shock (HSP) and cell cycle proteins, our analysis identified new differentially regulated proteins, including IL-32, MORF family mortality factors and apoptosis inducing factor SIVA, a target of p53. It could explain why proteasome inhibitors induce stronger apoptotic responses in immature AML cells.

Identification of potential serum biomarkers for breast cancer using a functional proteomics technology

BACKGROUND

Cancer is a genetic disease; its development and metastasis depend on the function of many proteins. Human serum contains thousands of proteins; it is a window for the homeostasis of individual's health. Many of the proteins found in the human serum could be potential biomarkers for cancer early detection and drug efficacy evaluation.

METHODS

In this study, a functional proteomics technology was used to systematically monitor metabolic enzyme and protease activities from resolved serum proteins produced by a modified 2-D gel separation and subsequent Protein Elution Plate, a method collectively called PEP. All the experiments were repeated at least twice to ensure the validity of the findings.

RESULTS

For the first time, significant differences were found between breast cancer patient serum and normal serum in two families of enzymes known to be involved in cancer development and metastasis: metabolic enzymes and proteases. Multiple enzyme species were identified in the serum assayed directly or after enrichment. Both qualitative and quantitative differences in the metabolic enzyme and protease activity were detected between breast cancer patient and control group, providing excellent biomarker candidates for breast cancer diagnosis and drug development.

CONCLUSIONS

This study identified several potential functional protein biomarkers from breast cancer patient serum. It also demonstrated that the functional proteomics technology, PEP, can be applied to the analysis of any functional proteins in human serum which contains thousands of proteins. The study indicated that the functional domain of the human serum could be unlocked with the PEP technology, pointing to a novel alternative for the development of diagnosis biomarkers for breast cancer and other diseases.

Opportunities and Challenges of Multiplex Assays: A Machine Learning Perspective

Multiplex assays that allow the simultaneous measurement of multiple analytes in small sample quantities have developed into a widely used technology. Their implementation spans across multiple assay systems and can provide readouts of similar quality as the respective single-plex measures, albeit at far higher throughput. Multiplex assay systems are therefore an important element for biomarker discovery and development strategies but analysis of the derived data can face substantial challenges that may limit the possibility of identifying meaningful biological markers. This chapter gives an overview of opportunities and challenges of multiplexed biomarker analysis, in particular from the perspective of machine learning aimed at identification of predictive biological signatures.

Mass-spectrometric exploration of proteome structure and function

Numerous biological processes are concurrently and coordinately active in every living cell. Each of them encompasses synthetic, catalytic and regulatory functions that are, almost always, carried out by proteins organized further into higher-order structures and networks. For decades, the structures and functions of selected proteins have been studied using biochemical and biophysical methods. However, the properties and behaviour of the proteome as an integrated system have largely remained elusive. Powerful mass-spectrometry-based technologies now provide unprecedented insights into the composition, structure, function and control of the proteome, shedding light on complex biological processes and phenotypes.

Reproducible quantitative proteotype data matrices for systems biology

Historically, many mass spectrometry–based proteomic studies have aimed at compiling an inventory of protein compounds present in a biological sample, with the long-term objective of creating a proteome map of a species. However, to answer fundamental questions about the behavior of biological systems at the protein level, accurate and unbiased quantitative data are required in addition to a list of all protein components. Fueled by advances in mass spectrometry, the proteomics field has thus recently shifted focus toward the reproducible quantification of proteins across a large number of biological samples. This provides the foundation to move away from pure enumeration of identified proteins toward quantitative matrices of many proteins measured across multiple samples. It is argued here that data matrices consisting of highly reproducible, quantitative, and unbiased proteomic measurements across a high number of conditions, referred to here as quantitative proteotype maps, will become the fundamental currency in the field and provide the starting point for downstream biological analysis. Such proteotype data matrices, for example, are generated by the measurement of large patient cohorts, time series, or multiple experimental perturbations. They are expected to have a large effect on systems biology and personalized medicine approaches that investigate the dynamic behavior of biological systems across multiple perturbations, time points, and individuals.

Integrative analysis of the colorectal cancer proteome: potential clinical impact

INTRODUCTION

Colorectal cancer (CRC) is one of the common types of cancer that affects a significant proportion of the population and is a major contributor to cancer related mortality. The relatively poor survival rate of CRC could be improved through the identification of clinically useful biomarkers. Areas covered: This review highlights the need for biomarkers and discusses recent proteomics discoveries in the aspects of CRC clinical practice including diagnosis, prognosis, therapy, screening and molecular pathological epidemiology (MPE). Studies have been evaluated in relation to biomarker target, methodology, sample selection, limitations, and potential impact. Finally, the progress in proteomic approaches is briefly discussed and the main difficulties facing the translation of proteomics biomarkers into the clinical practice are highlighted. Expert commentary: The establishment of specific guidelines, best practice recommendations and the improvement in proteomic strategies will significantly improve the prospects for developing clinically useful biomarkers.

Advances in Biomarkers: Going Beyond the Carcinoembryonic Antigen

Using biologically available markers to guide treatment decisions in colorectal cancer care is becoming increasingly common, though our understanding of these biomarkers is in its infancy. In this article, we will discuss how this area is rapidly changing, review important biomarkers being used currently, and explain how the results influence clinical decision-making. We will also briefly discuss the possibility of a liquid biopsy and explore several exciting and new options.

TRIC: an automated alignment strategy for reproducible protein quantification in targeted proteomics

Large scale, quantitative proteomic studies have become essential for the analysis of clinical cohorts, large perturbation experiments and systems biology studies. While next-generation mass spectrometric techniques such as SWATH-MS have substantially increased throughput and reproducibility, ensuring consistent quantification of thousands of peptide analytes across multiple LC-MS/MS runs remains a challenging and laborious manual process. To produce highly consistent and quantitatively accurate proteomics data matrices in an automated fashion, we have developed the TRIC software which utilizes fragment ion data to perform cross-run alignment, consistent peak-picking and quantification for high throughput targeted proteomics. TRIC uses a graph-based alignment strategy based on non-linear retention time correction to integrate peak elution information from all LC-MS/MS runs acquired in a study. When compared to state-of-the-art SWATH-MS data analysis, the algorithm was able to reduce the identification error by more than 3-fold at constant recall, while correcting for highly non-linear chromatographic effects. On a pulsed-SILAC experiment performed on human induced pluripotent stem (iPS) cells, TRIC was able to automatically align and quantify thousands of light and heavy isotopic peak groups and substantially increased the quantitative completeness and biological information in the data, providing insights into protein dynamics of iPS cells. Overall, this study demonstrates the importance of consistent quantification in highly challenging experimental setups, and proposes an algorithm to automate this task, constituting the last missing piece in a pipeline for automated analysis of massively parallel targeted proteomics datasets.

Advances in targeted proteomics and applications to biomedical research

Targeted proteomics technique has emerged as a powerful protein quantification tool in systems biology, biomedical research, and increasing for clinical applications. The most widely used targeted proteomics approach, selected reaction monitoring (SRM), also known as multiple reaction monitoring (MRM), can be used for quantification of cellular signaling networks and preclinical verification of candidate protein biomarkers. As an extension to our previous review on advances in SRM sensitivity (Shi et al., Proteomics, 12, 1074-1092, 2012) herein we review recent advances in the method and technology for further enhancing SRM sensitivity (from 2012 to present), and highlighting its broad biomedical applications in human bodily fluids, tissue and cell lines. Furthermore, we also review two recently introduced targeted proteomics approaches, parallel reaction monitoring (PRM) and data-independent acquisition (DIA) with targeted data extraction on fast scanning high-resolution accurate-mass (HR/AM) instruments. Such HR/AM targeted quantification with monitoring all target product ions addresses SRM limitations effectively in specificity and multiplexing; whereas when compared to SRM, PRM and DIA are still in the infancy with a limited number of applications. Thus, for HR/AM targeted quantification we focus our discussion on method development, data processing and analysis, and its advantages and limitations in targeted proteomics. Finally, general perspectives on the potential of achieving both high sensitivity and high sample throughput for large-scale quantification of hundreds of target proteins are discussed.

Mass Spectrometry Applied to Bottom-Up Proteomics: Entering the High-Throughput Era for Hypothesis Testing

Proteins constitute a key class of molecular components that perform essential biochemical reactions in living cells. Whether the aim is to extensively characterize a given protein or to perform high-throughput qualitative and quantitative analysis of the proteome content of a sample, liquid chromatography coupled to tandem mass spectrometry has become the technology of choice. In this review, we summarize the current state of mass spectrometry applied to bottom-up proteomics, the approach that focuses on analyzing peptides obtained from proteolytic digestion of proteins. With the recent advances in instrumentation and methodology, we show that the field is moving away from providing qualitative identification of long lists of proteins to delivering highly consistent and accurate quantification values for large numbers of proteins across large numbers of samples. We believe that this shift will have a profound impact for the field of proteomics and life science research in general.

Non-invasive prognostic protein biomarker signatures associated with colorectal cancer

The current management of colorectal cancer (CRC) would greatly benefit from non-invasive prognostic biomarkers indicative of clinicopathological tumor characteristics. Here, we employed targeted proteomic profiling of 80 glycoprotein biomarker candidates across plasma samples of a well-annotated patient cohort with comprehensive CRC characteristics. Clinical data included 8-year overall survival, tumor staging, histological grading, regional localization, and molecular tumor characteristics. The acquired quantitative proteomic dataset was subjected to the development of biomarker signatures predicting prognostic clinical endpoints. Protein candidates were selected into the signatures based on significance testing and a stepwise protein selection, each within 10-fold cross-validation. A six-protein biomarker signature of patient outcome could predict survival beyond clinical stage and was able to stratify patients into groups of better and worse prognosis. We further evaluated the performance of the signature on the mRNA level and assessed its prognostic value in the context of previously published transcriptional signatures. Additional signatures predicting regional tumor localization and disease dissemination were also identified. The integration of rich clinical data, quantitative proteomic technologies, and tailored computational modeling facilitated the characterization of these signatures in patient circulation. These findings highlight the value of a simultaneous assessment of important prognostic disease characteristics within a single measurement.