I'm working on a project in a high school bio lab (so limited resources), and I need a way to quantify the concentration of DNA in a PCR product. I can't use spectrophotometry (cheap spectrophotometers at school can't measure absorbance at 260nm), and any fluorescent dyes are out of the question as we don't have a transilluminator. Right now I'm looking at measuring absorbance change in a visible-light dye like crystal violet or methylene blue, but there isn't enough data on quantification using these dyes. Can anyone suggest anything?


  • $\begingroup$ Can you run a gel? $\endgroup$
    – canadianer
    Dec 11, 2017 at 0:20
  • $\begingroup$ Yes, although gel extractions might be a little difficult. $\endgroup$ Dec 11, 2017 at 0:51
  • $\begingroup$ If you are lucky, and you don't need many repeated measurements, and there is a university in your town, you can contact the administrative staff (note: not professors) of a fitting department (e.g.: genetics), who will usually put you in contact with a grad student of an experimental lab. Usually people are happy to help with "minor issues" like those - and having someone from a high school briefly visiting a lab will be good for any involved party, as it is a nice sign of community outreach. $\endgroup$
    – tsttst
    Dec 11, 2017 at 1:12
  • $\begingroup$ I do know a guy at a nearby uni who might be willing to help, but because the project is officially graded I'd need a pretty good set of data with replicates so it would mean basically doing everything at the uni, which I'm trying to avoid for now. I'll keep it in mind for a backup option, though. $\endgroup$ Dec 11, 2017 at 1:21
  • $\begingroup$ You can quantify DNA in a gel by comparing band intensity of the unknown sample to that of a known mass. I can elaborate if this seems like a reasonable strategy for you. $\endgroup$
    – canadianer
    Dec 11, 2017 at 6:57

1 Answer 1


Within some linear range, the intensity of stained DNA bands in a gel is directly proportional to their mass. By making a standard curve from bands of known mass, the mass of unknown bands can be estimated. This process is called densitometry. I wrote briefly about its use in quantifying protein bands after SDS-PAGE in this answer, but I'll go into more detail on how this is done using the following ethidium bromide stained gel:

enter image description here

On the right is a molecular weight marker (BstEII digest of λ DNA) with bands of known mass. On the left are two bands of unknown mass (actually a double-cut plasmid).

I use a free program called ImageJ to estimate band intensities. They have a tutorial on measuring band intensity here. You can also get ImageJ packaged with Java (and bunch of other stuff) in the software Fiji.

Open the image in ImageJ and use the Rectangle tool to draw a box around the marker, then select Analyze > Gels > Select First Lane (or shortcut Ctrl+1). Next, drag the box over to surround the other lane and select Analyze > Gels > Select Next Lane (Ctrl+2). Now select Analyze > Gels > Plot Lanes (Ctrl+3) to plot image intensity as a function of position on the gel:

enter image description here

The area under each peak (minus the background noise) represents the some total of intensity for each band. To measure this area, use the line (Straight) tool to draw bounds on the base of each peak:

enter image description here

I've done it somewhat haphazardly here, but you get the idea. Now, using the wand tool, click inside each peak and ImageJ will tabulate the integrated intensity of each. Shown below are the intensities of the marker bands (with the top band being Band 1) as well as their masses (which are known):

          Mass (ng)    Intensity
Band 1    45           5319.891
Band 2    28           3816.234
Band 3    24           3256.406
Band 4    16           2298.749
Band 5    15           2074.749
Band 6    8            1084.991

Take this data into Excel (or similar) and plot intensity as a function of mass. Band 1 looks like it is outside the linear range and is underestimating the mass. This was actually expected since, if you look closely at the band in the gel image, there are red pixels indicating that it is oversaturated. Ignore this band and perform a linear regression on the rest.

enter image description here

The regression gives the equation $I = 134.74M + 53.988$ with $R^2 = 0.9975$. Rearranging for mass, we have $M = \frac{I - 53.988}{134.74}$. Get the intensity of the unknown bands as described above and calculate their mass with the preceding equation:

          Mass (ng)    Intensity
Band 1    27           3736.991
Band 2    15           2125.284
  • 1
    $\begingroup$ My experience is that this is a much safer method than using (for example) Nanodrop. I have to admit though, that I rarely ever did the quantification in ImageJ/FIJI; estimating by eye was usually close enough for most applications :) $\endgroup$
    – gaspanic
    Jun 6, 2022 at 0:35

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .