Please use this identifier to cite or link to this item:
Title: Aplicación web para el estudio de relación entre Lupus y otras enfermedades autoinmunes basado en minería de textos de Pubmed
Author: Mendes Novo, Johanna
Tutor: Rebrij, Romina
Others: Merino, David  
Keywords: autoimmune
data mining
Issue Date: Jan-2020
Publisher: Universitat Oberta de Catalunya (UOC)
Abstract: The increase in publications in the biomedical literature it's a challenge for specialists to be up to date in their field of study or to summarize the information contained in thousands of publications. In this work, text mining is applied to Pubmed's abstracts, delivering an interactive web application in order to facilitate the use and distribution of the tool. The theme of autoimmune diseases (and especially the case of Lupus erythematosus) has been chosen since together, they affect more than 10% of the world's population. The iMune application allows us to explore the information contained in Pubmed's abstracts (a source with more than 30 million citations in the biomedical literature), search for publications on a disease of interest and summarize its content in word and gene clouds. For the particular case of Lupus, it allows us to explore potentially related genes by distinguishing those already mentioned in the reference literature. Gene-disease relationships are studied using latent semantic analysis. In addition, for the case of symptoms, the relationship of the most commonly presented symptoms in Lupus patients to other autoimmune diseases is explored. The tool has been able to summarize the words and genes contained in publications on autoimmune diseases, as well as to suggest related genes that can be tested in the literature or otherwise suggest new knowledge. Also, for a symptom, it exemplifies the complexity of diagnosis by verifying that a symptom is presented by a variety of autoimmune diseases
Language: Spanish
Appears in Collections:Trabajos finales de carrera, trabajos de investigación, etc.

Files in This Item:
File Description SizeFormat 
jmendesnTFM0120memoria.pdfMemoria del TFM2,81 MBAdobe PDFThumbnail