You are here : Home > Scientific news > ProteoRE, a web application for the discovery of biomarkers of diagnostic interest

Highlight | Proteomics | Bioinformatic | Biomarkers

ProteoRE, a web application for the discovery of biomarkers of diagnostic interest



Researchers at our institute have developed ProteoRE, a web application that provides a set of tools accessible to biologists to define their own strategy for selecting biomarker candidates.

Published on 20 November 2019
A biomarker is a precisely measurable characteristic used as an indicator of body function, disease or drug action. For example, blood glucose is a recognized biomarker both for characterizing diabetes and for evaluating the effectiveness of anti-diabetic molecules. In the medical field, a biomarker can be used for screening, diagnosis, response to treatment, relapse after treatment, etc. A biomarker can be a molecule (metabolites, circulating DNA, proteins, etc.) that can be easily measured in the blood or urine and is compatible with routine analysis. The development of a biomarker, from discovery to validation, is a long and costly process: how to accelerate the process of discovering new biomarker candidates for evaluation in clinical research?

Although biomarkers are widely used, limitations have been identified due to their lack of specificity; for example, in oncology, the absence of a single highly specific and sensitive marker has led to a growing consensus on the need for multiparameter biomarker combinations. An a priori knowledge-based approach (i.e. based on mechanistic criteria specific to the genesis of a disease) that takes advantage of ever-increasing "omics" databases could be a powerful way to identify biomarkers, provided that scientists have access to computer tools to search the data even when they have little experience in programming or bioinformatics support.
Based on this observation, researchers in our institute developed ProteoRE, a web application that provides a set of tools accessible to biologists to define their own strategy for selecting biomarker candidates. These tools are based on the extraction of experimental data from public human databases (proteomics and transcriptomics) and the application of successive filters based on physiological and biochemical criteria (e.g. tissue specificity, subcellular location, detectability, etc.) in relation to the pathology. A step-by-step protocol (i.e. a workflow) allows these tools to be linked in a transparent and reproducible way (Figure). The execution of this type of workflow via ProteoRE has thus enabled the researchers to select specific candidates of the cardiac muscle, located in the cytoplasm of the cells, as detectable and potential markers of lesions in the cardiac tissue [1]. In another study aimed to expand the list of biomarkers for early diagnosis of pancreatic cancer, this strategy resulted in a list of 14 biomarkers [2], some of which were supported by studies in patients diagnosed at an early stage - for subsequent evaluation by targeted proteomics

The bioinformatics tools, workflows, results and tutorials resulting from these strategies are made available to researchers (http://www.proteore.org/) so that they can be reused in their quest to discover biomarkers.


The ProteoRE project (Proteomics Research Environment) is a national project funded by the French Institute of Bioinformatics (IFB - ANR-11-INBS-0013), in collaboration with INRA (MIGALE platform) and carried out in cooperation with the proteomic infrastructure (ProFI). ProteoRE is a Galaxy-based web application open to the community and dedicated to the functional analysis and exploration of proteomic and transcriptomic data in biomedical research. This environment allows non-bioinformatics researchers to apply a wide range of tools and analysis workflows to their data, share their results with their collaborators, and reproduce the same analysis while keeping track of the overall process. Currently, ProteoRE includes 20 tools for data manipulation, annotation, classification, enrichment analysis and biological network construction, with associated visualizations.
"Omics" approaches are characterized as high throughput techniques allowing simultaneous analysis of a large number of variables, they mainly include: genomics, transcriptomics (gene expression and regulation), proteomics (protein analysis), metabolomics (study of produced metabolites). These approaches provide a wealth of information on the cellular and/or tissue response to in vitro or in vivo exposure.

Top page