Information Visualization to Enhance Sensitivity and Selectivity in Biosensing

• Grant number: 12/16802-6
• Support Opportunities: Regular Research Grants - Publications - Scientific article
• Start date: September 01, 2012
• End date: February 28, 2013
• Field of knowledge: Physical Sciences and Mathematics - Physics - Condensed Matter Physics
• Principal Investigator: Osvaldo Novais de Oliveira Junior
• Grantee: Osvaldo Novais de Oliveira Junior
• Host Institution: Instituto de Física de São Carlos (IFSC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

An overview is provided of the various methods for analyzing biosensing data, with emphasis on information visualization approaches such as multidimensional projection techniques. Emphasis is placed on the importance of data analysis methods, with a description of traditional techniques, including the advantages and limitations of linear and non-linear methods to generate layouts that emphasize similarity/dissimilarity relationships among data instances. Particularly important are recent methods that allow processing highdimensionaldata, thus taking full advantage of the capabilities of modern equipment. In this area, now referred to as e-science, the choice of appropriate data analysis methods is crucial to enhance the sensitivity and selectivity of sensors and biosensors. Two types of systems deserving attention in this context are electronic noses and electronic tongues, which are made of sensor arrays whose electrical or electrochemical responses are combined to provide "finger print" information for aromas and tastes. Examples will also be given of unprecedented detection of tropical diseases, made possible with the use of multidimensional projection techniques. Furthermore, ways of using these techniques along with other information visualization methodsto optimize biosensors will be discussed. (AU)