4.1.4 Results

The previously method of analysis has been applied to several molecules of 6.8 and 2.2 kbp. In particular, the FDC of 6 molecules of each length have been measured and analyzed. Figure 4.5 shows complete FDC of some of them. Since the unzipping curves are obtained at very low pulling rates, the process is quasistatic and the FDCs resemble each other. This is quite relevant because it indicates that the position and the size of the CURs are reproducible and they only depend on the sequence, emphasizing the reliability of the approach.

Figure 4.5: Experimental FDC of 3 molecules corresponding to the 2.2 kbp sequence (upper panel) and the 6.8 kbp sequence (lower panel). Since the raw data is too noisy, the data has been filtered with a low-pass running-average filter with a bandwidth of 1 Hz to clearly see the traces.

The histograms of metastable states are also very similar among them because they are obtained from FDCs that resemble each other. Figure 4.6 shows some of the obtained histograms for the 6.8 kbp sequence. Finally, Fig. 4.7 shows the average histograms and the Gaussian fits for the 2.2 and the 6.8 kbp sequences.

The resulting distribution of CUR sizes will be discussed in Sec. 4.3 together with the theoretically predicted histograms.

Figure 4.6: Histograms of intermediate states for six different molecules of 6.8 kbp. They are depicted in red, green, blue, magenta, orange and dark green. Although the height of each peak is different for the six histograms, the position of the peak is almost the same ($ \pm 10$ bp). The histograms for the 2.2 kbp molecules have similar reproducibility.

Figure 4.7: CUR size distributions. (a) Histogram of the intermediate states for the 2.2 kbp DNA sequence. Blue curve shows the experimentally measured histogram. Red curve shows the fit to a sum of Gaussians. (b) Histogram of intermediate states for the 6.8 kbp sequence. Same color code as in panel a.

JM Huguet 2014-02-12