How to use and integrate bioinformatics tools to compare proteomic data from distinct conditions? A tutorial using the pathological similarities between Aortic Valve Stenosis and Coronary Artery Disease as a case-study

• Type: Article in International Scientific Journal
• Date: 2018

Authors

Trindade, F (Author) Other; Ferreira, R (Author) Other; Magalhaes, B (Author) Other; Leite-Moreira AF (Author) FMUP; Falcao Pires, I (Author) FMUP; Vitorino, R (Author) FMUP

Journal

Title: Journal of Proteomics

Vol. 171

Pages: 37-52

ISSN: 1874-3919

Publisher: Elsevier

Indexing

ISI Web of Knowledge - 6 Citations

Scopus - 10 Citations

Other information

Authenticus ID: P-00N-FF5

DOI: 10.1016/j.jprot.2017.03.015

Abstract (EN): Nowadays we are surrounded by a plethora of bioinformatics tools, powerful enough to deal with the large amounts of data arising from proteomic studies, but whose application is sometimes hard to find. Therefore, we used a specific clinical problem-to discriminate pathophysiology and potential biomarkers between two similar cardiovascular diseases, aortic valve stenosis (AVS) and coronary artery disease (CAD) - to make a step-by-step guide through four bioinformatics tools: STRING, DisGeNET, Cytoscape and ClueGO. Proteome data was collected from articles available on PubMed centered on proteomic studies enrolling subjects with AVS or CAD. Through the analysis of gene ontology provided by STRING and ClueGO we could find specific biological phenomena associated with AVS, such as down-regulation of elastic fiber assembly, and with CAD, such as up-regulation of plasminogen activation. Moreover, through Cytoscape and DisGeNET we could pinpoint surrogate markers either for AVS (e.g. popeye domain containing protein 2 and 28S ribosomal protein S36, mitochondrial) or for CAD (e.g. ankyrin repeat and SOCS box protein 7) which deserve future validation. Data recycling and integration as well as research orientation are among the main advantages of resorting to bioinformatics analysis, hence these tutorials can be of great convenience for proteomics investigators. Biological significance: As we saw for aortic valve stenosis and coronary artery disease, it can be of great relevance to perform preliminary bioinformatics analysis with already published proteomics data. It not only saves us time in the lab (avoiding work duplication) as it points out new hypothesis to explain the phenotypical presentation of the diseases as well as new surrogate markers with clinical relevance, deserving future scrutiny. These essential steps can be easily overcome if one follows the steps proposed in our tutorial for STRING, DisGeNET, Cytoscape and ClueGO utilization.

Language: English

Type (Professor's evaluation): Scientific

No. of pages: 16