Were all of your measurements contained within the dynamic range of your standard curve? Extrapolation beyond the standard curve or measurements made close to the limit of detection can introduce significant error.

I would recommend Adam's reply, you should take into consideration only the experimental values that hit the middle of your standard curve, error being more prominent in either low or high end of the curve.

In addition I would suspect a biological effect from the urine sample. I suspect that the standard used in the assay is a purified protein and not "neat" urine with specified amount of your analyte. Therefore your standard and your analyte are not the same. Biological background is a typical technical issue at >10-20% sample concentrations and it dilutes out. If possible I would always use less than 5% (1/20 diluted) sample concentration to factor it out.

I agree with Adam. Check the lower limit of detection and lower limit of quantification for your assay. As I have found out, at the extremes of the standard curve, although you can detect sample it is not always quantifiable.

Thank you all for your answers!

Adam B Shapiro 2 of the 4 were too highly concentrated in their neat sample but after 1:2 dilution they both came within range. The other 2 samples sit in the middle of the range up to 1:8 dilution.

Juha Kotimaa - you are correct the standard is a recombinant version of the protein as we are looking for an endogenous protein, I know this is a general problem for biomarker ligand binding assays as they are not identical. I'm just wondering what type of interference we might be seeing that could cause a loss of recovery?

Hi Chloe,

Well, simply put you might see elevated background. You state:

" Donor A neat urine measured 2500 pg/mL and the 1:8 dilution measured 1200 pg/mL after dilution correction (75/0.0625 = 1200)."

Without seeing the raw data (and the standard curve), I could throw out a guess. If both values are extrapolated from a decent standard curve and within the linear range, one of the values is either over, or understated.

So, guestimating that the biological background (or matrix effect) increases the raw signal, the neat measurement is overstated and the diluted sample should be closer the truth.

Say you get a raw signal of 0.787 from the neat urine sample, but in reality urine without your analyte could give a signal of 0.350. When you take that into account you will end up with a value of 0.787-0.350 = 0.437. Similarly, you would have lower absolute amount of protein after calculating through the standard curve.

Juha Kotimaa That makes sense to me now, thank you for explaining! How would you determine if the neat urine is overstated? I am looking at an endogenous protein so it is difficult to obtain urine without the analyte in order to compare.

Chloe Guest That's a bit of a problem for sure. One option would be to carefully compare the standard curve and the dose response curve you receive from various urine samples.. If the trend lines of the linear ranges of the slopes are radically different, the standard won't give you an accurate reflection of amount of protein in the urine sample. However there ought to be a range where the effect of the matrix dilutes out. When calculating the pg/ml values of the urine samples analysed in dose response, there should be some consecutive dilutions which will give you close matching results. This is the range where the matrix effect should be negligible and give the answer at which dilution range the assay works with urine.

You could always send the data to the manufacturer and check if they have any suggestions as well.

Juha Kotimaa Unfortunately I think the endogenous levels are too low for us to reach a point where the matrix effect is corrected and also the concentrations are within the assay range - I think this is a common problem with endogenous compounds from the reading I've been doing. Thank you again for your help, it's been very useful!

Can you concentrate the protein of interest in the urine by ultrafiltration, or dialyze the urine to reduce the background?