Resumo

GALVEZ, Carmen. Searching for drug interactions in MEDLINE with the use of measures of centrality. Rev. cuba. inf. cienc. salud [online]. 2016, vol.27, n.2, pp. 154-167. ISSN 2307-2113.

The paper proposes an approach to the identification and prediction of drug interactions in biomedical literature using measures of centrality. Drug interactions are caused by alterations in the effect of a drug. Health specialists have pharmacological databases at their disposal in which information is provided about such interactions. However, because such databases have a limited scope, biomedical literature continues to be the source of scientific information par excellence. The method used to identify such interactions was based on network analysis and information visualization techniques. Degree, closeness and betweenness metrics were applied to a set of drugs extracted from the database MEDLINE with the purpose of classifying the drugs in the network. The results obtained show that the centrality of betweenness is the most appropriate measure to identify and predict new interactions. The conclusion is that the drug interactions revealed by the procedure proposed could be good candidates for further experimental analysis aimed at verifying their clinical relevance. The procedure could also be used for the content curation of drug databases.

Palavras-chave : network analysis; measures of centrality; drug interactions.

        · resumo em Espanhol     · texto em Espanhol     · Espanhol ( pdf )