Use of Information Visualization Methods Eliminating Cross Talk in Multiple Sensing Units Investigated for a Light-Addressable Potentiometric Sensor

摘要

The integration of nanostructured films containing biomoleculesrnand silicon-based technologies is a promising directionrnfor reaching miniaturized biosensors that exhibit highrnsensitivity and selectivity. A challenge, however, is to avoidrncross talk among sensing units in an array with multiplernsensors located on a small area. In this letter, we describernan array of 16 sensing units of a light-addressable potentiometricrnsensor (LAPS), which was made with layer-by-layerrn(LbL) films of a poly(amidomine) dendrimer (PAMAM) andrnsingle-walled carbon nanotubes (SWNTs), coated with arnlayer of the enzyme penicillinase. A visual inspection of therndata from constant-current measurements with liquidrnsamples containing distinct concentrations of penicillin,rnglucose, or a buffer indicated a possible cross talk betweenrnunits that contained penicillinase and those that did not.rnWith the use of multidimensional data projection techniques,rnnormally employed in information visualizationrnmethods, we managed to distinguish the results from thernmodified LAPS, even in cases where the units were adjacentrnto each other. Furthermore, the plots generated with therninteractive document map (IDMAP) projection techniquernenabled the distinction of the different concentrations ofrnpenicillin, from 5 mmol L-1 down to 0.5 mmol L-1. Datarnvisualization also confirmed the enhanced performancernof the sensing units containing carbon nanotubes, consistentrnwith the analysis of results for LAPS sensors. Thernuse of visual analytics, as with projection methods, mayrnbe essential to handle a large amount of data generatedrnin multiple sensor arrays to achieve high performancernin miniaturized systems.