Abordagens da biologia de sistemas na investigação dos pontos de articulação nas rotas metabólicas do KEGG

• Ano de defesa: 2020
• Autor(a) principal: Brandão, Igor Augusto
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Rio Grande do Norte; Brasil; UFRN; PROGRAMA DE PÓS-GRADUAÇÃO EM BIOINFORMÁTICA
• Palavras-chave em Português: Biologia de sistemas; KEGG; Ponto de articulação; Via metabólica; Topologia de redes; Visualização de redes.; Systems biology; Articulation point; Metabolic pathway; Networktopology; network visualization
• Link de acesso: https://repositorio.ufrn.br/handle/123456789/30737

The study of proteins essentiality through laboratory methods is expensive, time-consuming and not scalable for large amounts of proteins. Besides, it is relevantto evaluate the essentiality of several proteins of a metabolic pathway as a whole. Themetabolic pathways can be analyzed as graphs, which provide several tools to studythe topological features such as the articulation points. Nowadays, research in bioinfor-matics studies the essentiality of proteins based on betweenness and degree metrics,however, graph theory suggests that articulation points could be essential nodes in anetwork. It remains to be determined whether these articulation points are essential inmetabolic pathways and their topological impact on the network. Using network analysisvia metrics and biologic curation, we aim to verify if bottlenecks are proteins with thehighest frequencies and located in the center of KEGG metabolic pathways. For thispurpose, we identified the articulation points in different networks, evaluate the impactof each articulation point, calculate their frequency and compare them with occurrencesof non-articulation points. We consulted KEGG pathways available as KGML files. After,the data was transformed into a graph object. Two centrality parameters includingarticulation points and degree are determined and the essential proteins based onthese parameters are classified. Approximately 20% of the proteins are articulationpoints. The articulation points with high-frequency which are located in central regionsof the network were considered the most important (3.75%). In addition, the highestconcentration of articulation points occurred in the frequency range of 80-90%. A patternof non-randomness of articulation points was identified in the protein groups that havea frequency of at least 74.5%. Finally, steroid biosynthesis is the metabolic pathwaywith the highest number of articulation points with frequency higher than 80%. Besides,oxidoreductase is the articulation point class present in the highest number of metabolicpathways. Overall, the findings suggest that bottlenecks are articulation points withhighest frequencies and located in the center of the network. It remains to perform adeep analysis on the articulation points biological roles