[perivar]

Hello. i have some question regarding use of bradford method.
 
why is linerization of the bradford method more uncertain near 0 mg/mL concentration of protein and above 0,9 mg/mL where it bends off. In an experiment, the regressive line usually never goes by (0,0) which i assume has something to do with the uncertanty.
 
Using the UV - method give a more "perfect" regressive line with high precision, but if measuring contaminated protein, often gives an overestimation of the concentration, where the bradford method gives a more correct assumtion of the concentration but with lesser precision. Why is the precision lesser with bradford method?

[Arkcon]

Hello. i have some question regarding use of bradford method.

OK, and you've given us lots to do here:
 

why is linerization of the bradford method more uncertain near 0 mg/mL concentration of protein and above 0,9 mg/mL where it bends off.

I don't know how the "linerization" can be called more uncertain at one point of the line than another.  Maybe you should describe the process you used to determine the line.

In an experiment, the regressive line usually never goes by (0,0) which i assume has something to do with the uncertanty.

This is a common observation.  The reason you give is not the reason why, however.
 

Using the UV - method give a more "perfect" regressive line with high precision,

Great, you've used regression to determine the line.  And you've used a UV-Vis spec to quantify your assay.  That information helps, but ...

but if measuring contaminated protein, often gives an overestimation of the concentration,

You observed this?  How did you do it?

where the bradford method gives a more correct assumtion of the concentration but with lesser precision. Why is the precision lesser with bradford method?

OK, something important for a beginner to do is to carefully define for themselves the meanings of "accuracy" and "precision".  These two words are not synonyms.  Once defined, when we speak, we only use them when we mean them, and we abandon terms like "correct" and "better"

[Yggdrasil]

For high concentrations of protein, the Bradford assay is inaccurate because it becomes difficult to measure the absorbance of the solution.  At high absorbance values (>1.0 AU), such little light passes through the sample to the detector that you cannot reliably measure absorbance values.  At low protein concentrations, the Bradford assay is also inaccurate because contaminants (e.g. dust, which contains protein) can interfere.  These contaminants and the fact that a small amount of the Bradford dye populates the colored state at equilibrium explain why the absorbance of a sample without protein is not zero.

but if measuring contaminated protein, often gives an overestimation of the concentration,

You observed this?  How did you do it?

Nucleic acids absorb nearly the same wavelengths as protein (λmax = 260 for DNA/RNA, λmax = 280 for protein), so measurements of protein concentration using A280 are very sensitive to nucleic acid contamination.

It's also worth noting that the Bradford assay is sensitive to contamination by lipids or surfactants.

[Babcock_Hall]

The Bradford assay usually goes nonlinear in my hands at absorbances lower than the bicinchoninic acid (BCA) assay.  I seem to recall that this is because there is not a great excess of the dye in solution, and it begins to become depleted at high protein concentration.  I don't have a citation handy, but I might be able to turn one up if need be.  However, some commercial preparations of the Bradford reagent are advertised as being less prone to this problem.

Both of these assays show some protein-to-protein variation in sensitivity.  For example, bovine albumin gives about twice the response as egg albumin does in a Bradford, and this is pretty reproducible in my hands.

I agree that the absorbance of a Bradford assay solution at zero concentration of standard protein is nowhere near zero in absorbance.  I often observe it to be about 0.4 to 0.5 absorbance units, but I suspect that there are a number of factors which affect this number.  UV absorbance measurements will also pick up oxidized 2-mercaptoethanol (the disulfide form has λ_max ≈ 270-280 nm) and might pick up things like PMSF (I don't have its absorption spectrum in front of me), as well as nucleic acids and some cofactors.  Turbidity can also be a problem.

[Arkcon]

I agree that the absorbance of a Bradford assay solution at zero concentration of standard protein is nowhere near zero in absorbance.  I often observe it to be about 0.4 to 0.5 absorbance units, but I suspect that there are a number of factors which affect this number.  UV absorbance measurements will also pick up oxidized 2-mercaptoethanol (the disulfide form has λ_max ≈ 270-280 nm) and might pick up things like PMSF (I don't have its absorption spectrum in front of me), as well as nucleic acids and some cofactors.  Turbidity can also be a problem.

To summarize what Babcock_Hall: says here, spectrometer assays don't have to go through zero.  The blank mixture without protein, even though it doesn't appear to be the correct color still absorbs slightly.  Or scatters.  Whatever.  Again, as I said, this is not how experimental uncertainty is defined.

We've been discussing this topic, and the topic of non-linearity at high O.D. lately.  Try a search on this forum.