Genomic analysis of macrophage gene signatures during idiopathic pulmonary fibrosis development

• Autor(a) principal: Cruz, Giuliano Netto Flores
• Data de Publicação: 2018
• Tipo de documento: Trabalho de conclusão de curso
• Idioma: eng
• Título da fonte: Repositório Institucional da UFRGS
• Texto Completo: http://hdl.handle.net/10183/195681
• topic: Fibrose pulmonar idiopática; Macrófagos

Idiopathic Pulmonary Fibrosis (IPF) is a chronic, progressive, irreversible lung disease. After diagnosis, the interstitial condition commonly presents 3-5 years of life expectancy if untreated. Despite the limited capacity of recapitulating IPF, animal models have been useful for identifying related pathways relevant for drug discovery and diagnostic tools development. Using these techniques, several immune-related mechanisms have been implicated to IPF. For instance, subpopulations of macrophages and monocytes-derived cells are recognized as centrally active in pulmonary immunological processes. One of the most used technologies is high-throughput gene expression analysis, which has been available for almost two decades now. The “omics” revolution has presented major impacts on macrophage and pulmonary fibrosis research. The present study aims to investigate macrophage dynamics within the context of IPF at the transcriptomic level. Using publicly available gene-expression data, we applied modern data science approaches to (1) understand longitudinal profiles within IPF models; (2) investigate correlation between macrophage genomic dynamics and IPF development; and (3) apply longitudinal profiles uncovered through multivariate data analysis to the development of new sets of predictors able to classify IPF and control samples accordingly. Principal Component Analysis and Hierarchical Clustering showed that our pipeline was able to construct a complex set of biomarker candidates that together outperformed gene expression alone in separating treatment groups in an IPF animal model dataset. We further assessed the predictive performance of our candidates on publicly available gene expression data from IPF patients. Once again, the constructed biomarker candidates were significantly differentiated between IPF and control samples. The data presented in this work strongly suggest that longitudinal data analysis holds major unappreciated potentials for translational medicine research.