A schematic of the high-speed magnetic tweezer with a fiber-coupled superluminescent diode (SLD) for illumination and a high-speed CMOS (hs-CMOS) camera for detection. The inset shows a 256 × 256 8-bit brightfield-normalized still-frame, taken from a video acquired at 35 087 fps. The still frame shows a partially molten 2.5 μm diameter reference bead and a 1.05 μm diameter streptavadin-coated magnetic bead, which is tethered to the surface via a DNA hairpin.
Schematic illustration of data flow in our high-speed magnetic tweezer. Images from the hs-CMOS camera are transferred (with delay) via ethernet cable to the CPU. The CPU transfers the data to the GPU, where XYZ coordinates are generated.
Three ways to examine the instrumental tracking error of a magnetic tweezer. Figure 3(a) shows the position of a partially molten polystyrene reference bead as a function of time with (corrected) and without (raw) reference bead subtraction. Figure 3(b) shows the Allan deviation of bead position as a function of the measurement time, for lateral X and axial Z fluctuations. Figure 3(c) shows the PSD of bead motion, with and without reference bead subtraction. For clarity, lateral Y-fluctuation data is not shown in Figures 3(b) and 3(c) , but it qualitatively matches the X-fluctuation data.
Extension trajectory of a DNA hairpin undergoing stochastic folding/unfolding transitions. Data was acquired at 35087 fps, then low-pass filtered to 701.74 fps. Inset shows a 0.011 s residence in the folded state. The histogram of particle positions is well-described by a sum of two Gaussians, separated by 17.6 nm.
Article metrics loading...
Full text loading...