How purity (%) of protein can be determined by densitometry of coomassie brilliant blue stained SDS-PAGE gel ?
Scan the entire lane. Using the image analysis software, measure the total density of the band of the protein of interest by drawing a box or freehand shape around it. Then measure the total density of the entire lane the same way. Subtract the background density of a suitably matched area on the gel in each case. Divide the background-corrected density of the protein band by the background-corrected density of the whole lane and multiply by 100 to get % purity. (There is an assumption here that all proteins stain equally well.)
To do this properly, you should not allow any of the proteins in the sample to run off the gel. Use a high percentage gel, and include any density running at the dye front. Use a pretty heavy loading so that you can detect low-abundance bands.
Do not be surprised if the purity comes out lower than you think it should be based on visual appearance. People are not good at judging this by eye.
Impurities running together with the protein of interest will be counted as the protein of interest, so the purity will actually be an overestimate. Use a light loading (or preferably a 2-D gel) to see if there are any such proteins in a significant amount.
Scan the entire lane. Using the image analysis software, measure the total density of the band of the protein of interest by drawing a box or freehand shape around it. Then measure the total density of the entire lane the same way. Subtract the background density of a suitably matched area on the gel in each case. Divide the background-corrected density of the protein band by the background-corrected density of the whole lane and multiply by 100 to get % purity. (There is an assumption here that all proteins stain equally well.)
To do this properly, you should not allow any of the proteins in the sample to run off the gel. Use a high percentage gel, and include any density running at the dye front. Use a pretty heavy loading so that you can detect low-abundance bands.
Do not be surprised if the purity comes out lower than you think it should be based on visual appearance. People are not good at judging this by eye.
Impurities running together with the protein of interest will be counted as the protein of interest, so the purity will actually be an overestimate. Use a light loading (or preferably a 2-D gel) to see if there are any such proteins in a significant amount.
ImageJ (https://imagej.nih.gov/ij/) is a free program that can do what Adam suggested.
Just to amplify something Adam also mentioned: Not all proteins bind dyes equally well. With fluorescent dyes like RuBPS variability seems to be lower than with Coomassie. You also need less material and you may get a larger linear range.
I found this video link online when I was searching for the same question
https://www.youtube.com/watch?v=JlR5v-DsTds.
Also attaching a document in which there are instructions for ImageJ.
Hope it helps.
- Badges
- Science method