Whatever you are measuring, the first step to getting accurate results is an uncontaminated sample. Take steps to prevent contamination during preparation by wearing gloves, using only sterile containers, and changing your pipet tip after each use. Also take care not to contaminate your pipet by pulling your sample past the tip.
Many errors stem from problems with the gel. First, an incorrect gel concentration can result in samples running too quickly or not at all. Make sure you are using a concentration that is correctly measured and geared toward the general sizes of the sample that you are attempting to measure. For instance, when you look at a DNA sample, a 0.7 percent gel would allow you to effectively sort fragments that are anywhere from 800 to 10,000 base pairs long, whereas smaller fragments would necessitate a higher percentage gel.
Gel can be compromised in other ways. Before loading your samples, make sure the gel has no cuts or bubbles. Even a small imperfection can affect your results. Proper preparation of the gel is key. Follow all instructions, pour slowly to avoid bubbling, and let your gel cool completely before you load your samples. The molecular matrix of the gel itself and the ability of your sample to navigate through this matrix to the positively charged end of the tray are what demonstrate the sample's size. A physical disruption or even a temperature that is too extreme will affect this molecular matrix and compromise the validity of your data.
Finally, when setting your gel you will place a comb to create the wells where you will load your samples. Take care that this comb is placed properly in the insets of your tray, because a comb that creates wells that are too deep or shallow can lead to errors. The comb should not reach completely through the gel to the bottom of the tray, but should create wells deep enough to load your samples without overflowing.
Although it seems simple enough, loading your samples into the small wells in the gel can prove difficult. Most important, make sure you know how much to put in each well. Too much of your sample could cause it to smear or appear smaller than it truly is; too little of your sample may be impossible to see. Also, do not puncture the gel when loading your samples, because this can cause problems just like other gel imperfections.
A common mistake for students new to the gel electrophoresis technique is running the gel backwards. This happens when the positive and negative connections are attached to the wrong ends of the tray. You can avoid this mistake by remembering that the sample will always run to red, so make sure the positive end is attached on the opposite side of the tray from the sample wells.
Problems also occur with the use of incorrect voltage. A voltage that is too high can cause the sample to run off the other end of the gel, while one that is too low may mean the sample will take too long to run or never run far enough to be quantified usefully. Incorrect placement of the tray within the current or disruptions in the current may cause the sample to run diagonally or inconsistently, which can make accurate measurement difficult.
If you cannot see your results, you obviously will have trouble interpreting them. A sample that is too small can be difficult to visualize. Problems with the visualization methods can also compromise the results. The most common methods use visible dyes or ethidium bromide (for visualization under ultraviolet light). These must be added to the gel at the correct concentrations, or visualization may be hindered.
Varied measurement is a widely accepted source of error in gel electrophoresis that students often overlook. When you measure the length of gel that your sample traveled, your ruler will be limited in accuracy, likely to a millimeter. When your measurement appears to lie between millimeters, you will not be able to specify the exact measurement. Also, the bands formed by electrophoresis may vary in thickness. Did you measure to the top or bottom of the band in each column? Consistency is important, but measuring to the edge of the band farthest from the well is most common. Keep in mind that this source of error also likely affected your sample amount and concentrations of other components in your experiment.