A data-driven, knowledge-based approach to biomarker discovery: application to circulating microRNA markers of colorectal cancer prognosis

Abstract

Recent advances in high-throughput technologies have provided an unprecedented opportunity to identify molecular markers of disease processes. This plethora of complex-omics data has simultaneously complicated the problem of extracting meaningful molecular signatures and opened up new opportunities for more sophisticated integrative and holistic approaches. In this era, effective integration of data-driven and knowledge-based approaches for biomarker identification has been recognised as key to improving the identification of high-performance biomarkers, and necessary for translational applications. Here, we have evaluated the role of circulating microRNA as a means of predicting the prognosis of patients with colorectal cancer, which is the second leading cause of cancer-related death worldwide. We have developed a multi-objective optimisation method that effectively integrates a data-driven approach with the knowledge obtained from the microRNA-mediated regulatory network to identify robust plasma microRNA signatures which are reliable in terms of predictive power as well as functional relevance. The proposed multi-objective framework has the capacity to adjust for conflicting biomarker objectives and to incorporate heterogeneous information facilitating systems approaches to biomarker discovery. We have found a prognostic signature of colorectal cancer comprising 11 circulating microRNAs. The identified signature predicts the patients’ survival outcome and targets pathways underlying colorectal cancer progression. The altered expression of the identified microRNAs was confirmed in an independent public data set of plasma samples of patients in early stage vs advanced colorectal cancer. Furthermore, the generality of the proposed method was demonstrated across three publicly available miRNA data sets associated with biomarker studies in other diseases.

Differentiation of lymphocytic‐plasmacytic enteropathy and small cell lymphoma in cats using histology‐guided mass spectrometry

“…[20][21][22]27 Stages of biomarker discovery commonly involve the discovery and identification of biomarkers within a training set, and the validation of the assay using an independent set of samples. 28,29 In the initial stages of assay development, samples should be well characterized. In the light of a lacking gold standard for the diagnosis and differentiation of LPE from SCL in cats, this study used the decisions of a panel of experts consisting of board-certified anatomic pathologists, internists, and oncologists.…”

“…Stages of biomarker discovery commonly involve the discovery and identification of biomarkers within a training set, and the validation of the assay using an independent set of samples 28,29 . In the initial stages of assay development, samples should be well characterized.…”

The Translational Status of Cancer Liquid Biopsies

“…Systems biology provides a framework to investigate complex cancer phenotypes in terms of pathways and networks. The state-of-the-art statistical and computational algorithms can be applied to the accumulated multi-layered biological data and integrated with known cancer-related biochemical pathways to guide the discovery of new biomarker panels [434]. The biomarker candidates can then be analytically and clinically validated in clinical trials.…”

A novel deep mining model for effective knowledge discovery from omics data

“…The question is what are the required elements of a feature learning algorithm to be able to exploit large and noisy spaces of omics data effectively and discover robust biomarkers? Given the fact that omics data are more likely to be non-linear in nature [3], there is a necessity for nonlinear feature learning that avoids the linear assumptions of traditional statistical techniques in order to discover enough of the meaningful intricacies underlying these high-throughput biomedical data. Since it has been shown that learning models with a single stage of input transformation (i.e.…”