Depending on the data you collect, you may need to use a non-linear best fit equation. The DNA ladder on agarose gels and the Protein ladder on SDS-PAGE gels both run with a non-linear correlation to sample size. Use the following example agarose gel (Figure 1) to determine the best fit line. Follow the steps below the gel example.
Figure 1.
Use this example of an agarose gel to generate a non-linear standard curve equation to solve for the DNA bands in the lane for Sample A and Sample B. The written instructions are below the figure. The sizes for the marker/ladder are to the right.
Draw a horizontal line (in ballpoint pen or pencil so it doesn’t smudge) on your gel example through the vertical center of the wells. You will use this line as the starting point for all your measurements.
Then, use a ruler to measure (in mm) the distance from the center of the well to the center of each band. This tells you how far each band migrated.
Once you have your distance data for the marker lane, correlate each migration distance with the appropriate size in bp. (Which will have traveled farther: small fragments or big fragments?) The standard curve should include data ONLY from the marker/ladder lane.
Enter the marker lane data into an Excel spreadsheet, and make a scatter plot with a title and labeled axes as you did in the previous learning activity (Make a Standard Curve Graph in Excel). (Good axis labels: DNA fragment size (bp), Distance traveled (mm).
In Excel click on a the graph to open the three boxes to the right of the graph. Click on the one with a plus sign (+) and click on the arrow pointing right next to the option for trendline. Click More Options... to open the dialog box on the right (see figure 4).
Check the boxes to choose a Logarithmic trendline, and Display the R-squared value and Equation. Note that the equation and R-squared value now appear on your graph. Real gels sometimes show deviation from logarithmic migration. If power, polynomial, or some other curve gives a better R-squared value than logarithmic, you should use the curve that gives the best fit.
Solve for an x (bp) value given a y (migration in mm) value from a logarithmic function.
For a natural log (ln) best fit equation with the format y=m*ln(x)+b.