Accurate detection of sample location in isotachophoresis
Merav Karsenty, Tally Rosenfeld, Khaled Gommed, and Moran Bercovici
We developed a new method for accurate detection of sample location in peak mode isotachophoresis (ITP) in a microchannel containing multiple constrictions using electric current reading alones.
We developed an algorithm for detecting changes in electric current. The algorithm works in real time and indicates the exact time in which the ITP interface enters the constriction.
We developed an analytical model that could be used to predict the electric current curves for arbitrary constriction geometries and applied voltages, and thus assist in future optimizations for various purposes.
We demonstrate the use of the technique to deliver sample to a designated location in the channel with an accuracy as low as 50 μm.
Current monitoring in microchannel with constrictions
Video 1. Experimental results demonstrating the use of channel constrictions for accurate determination of ITP plug location. Electric current monotonically decreases as the low mobility TE displaces the high mobility LE. When the ITP interface enters the constriction, a rapid and significant drop in current occurs, resulting as a step pattern in the current signal.
Location detection algorithm
Video 2. Our algorithm is designed to detect these steps in electric current in real time, indicating the exact time of entrance into the constriction. We detect the transitions in real time by cross-correlation of the signal with a predefined step function. Maximum correlation signal is expected when the step function overlaps with the step in the current signal. Red circles indicate the points in time in which the algorithm detects the transition (i.e. local maxima in the cross-correlation). The method is robust to noise and peak values indicate entrance to the constriction regions. Peaks are recognized by the computer in real time and the ITP plug can be stopped at a designated location.
Delivery of focused sample
Video 3. A focused sample is delivered to a designated location. As the ITP interface enters a constriction, the automatic tracking algorithm detects the ITP plug location and turns off the electric field as it enters the designated chamber.
Sample delivery accuracy
Fig1. Experimental results demonstrating the accuracy of the technique. Blue circles represent the stop locations of five experiments, in which the algorithm was set to turn off the electric field after the ITP plug exits the second constriction. We define the stopping point as the location of the maximal signal of the width averaged intensity. Red dashed lines represent the range of 95% confidence on the mean location, and show a 50 μm accuracy in sample delivery.
Robustness to contamination
Fig 2. Experimental results demonstrating the robustness of the methods to significant contamination of the LE. (a) Electric current vs. time for ITP performed using LE spiked with NaOH concentrations between 0 and 50 mM. As expected, the absolute values of the current as well as its rate of change vary significantly between experiments. The electric current value thus cannot be used to robustly detect the location of the ITP interface. (b) Experimental results showing the automatically obtained stop locations, with the LE spiked with NaOH concentration between 0 and 50 mM (five repeats of pure LE case and three repeats for all other cases). Red dashed lines represent a 95% confidence on the mean, accounting for all cases and repeats. Results show the algorithm is insensitive to such contaminations and can deliver the focused sample to a desired location with an accuracy of 70 μm.