Plasmid ID by comparing individual experimental barcodes to a theory database.

<p>(Left) Pearson correlation coefficients between the experimental barcodes (<i>pUUH</i>, <i>pEC005A</i> and <i>pEC005B</i>) and the plasmid theory database (as in [<a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0179041#pone.0179041.ref024" target="_blank">24</a>]) were calculated and turned into histograms. As input experimental barcode we used: unfiltered single-time frame (<i>ust</i>) barcodes, filtered (with Gaussian filter) single-time frame (<i>fst</i>) barcodes, and time-averaged (<i>ta</i>) barcodes. The vertical line gives the correlation coefficient for the correct plasmid. Notice that <i>ust</i> barcodes are not very good at plasmid ID (s-scores between 0.4-10%), but, <i>fsb</i> and <i>ta</i> barcodes are better (with s-scores less than 3%). Experiments were only matched to theory barcodes which had a length within ±3<i>σ</i>_length of the experimental barcode, with <i>σ</i>_length = {23, 7, 15} pixels for <i>pUUH, pEC005A and pEC005B</i>, respectively. The Gumbel fits to the histograms were done using the same method as in [<a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0179041#pone.0179041.ref024" target="_blank">24</a>]. (Right) The panels on the right show the change in between the experiment and the theoretical barcode (i.e., theory barcodes from <i>pUUH</i>, <i>pEC005A</i> or <i>pEC005B</i>, respectively) after filtering single time frame barcodes. On the top-right plot, in grey is shown the correlation coefficients for the case without filtering, and in blue/green the correlation after filtering. The four lines in the top plot are smoothed for visualization purposes and are moving averages of the result over the 300 nearest neighbor barcodes (each of the three plasmid types treated separately). The right-bottom panel shows the change for all 6400 single time-frames for the case of Gaussian filter. On average, the single frames match to theory improves by 0.11 ± 0.04 points after filtering (average over all three filters).