Desenvolvimento de workflows científicos para a geração e análise de diferentes redes de interatomas

Abstract

Scientific workflows can be used to design in silico experiments to process and analyse a great amount of data using computational simulation. In this sense, all in silico experiments are organized as a sequence of steps that characterize an execution flow, where each steps employ different softwares. It is important to note that these softwares do not employ a common structure or model to represent data. Thus, taking into account the diversity of softwares and data representation, a scientific workflow should be built to integrate and process data, where each step must be able to analyse the data structure, to process these data and finally generate a new data structure that provide the requirements needed to execute the next step of the workflow. It is important to note that scientific workflows use mathematics and computer sciences methods capable to process, storage, and transform these data into useful information. In addition, a subarea of Bioinformatics termed Systems biology helps to understand the interactions observed between populations of molecules, cells and organisms as the result of the increase of biological complexity. Thus, the purpose of this work was to develop scientific workflows using bioinformatics and system biology softwares with the purpose to compare different biological processes from the generation of proteinprotein interaction (PPI) networks. In this sense, the Dis2PPI software was design as a scientific workflow to integrate OMIM (Online Mendelian Inheritance in Man) database with metasearch program STRING (Search Tool for the Retrieval of Interacting Genes/Proteins) in order to retrieve protein-protein interaction networks associated to monogenic diseases. The Dis2PPI was used to generate PPI networks for xeroderma pigmentosum and cockayne syndromes, two rare monogenic diseases associated with DNA damage, cancer, progeria, and neurodegeneration. The PPI networks generated were analyzed with Cytoscape program and specific plugins that evaluate gene ontologies and centrality analysis. By its turn, Net2Homology program was designed as a scientific workflow to integrate NCBI PSI-BLAST sequence alignments with STRING metasearch program to recovery PPI networks associate with two different species. In this sense, Net2Homology was used to obtain PPI network for xeroderma pigmentosum disease and cyclooxigenase-1 enzyme by crossing Homo sapiens and Mus musculus information. A comparison between these networks was realized to verify the major evolutively conserved biological process.