Prism can subtract nonspecific from total binding to determine specific binding at each concentration.

To subtract nonspecific binding:

1. Enter concentrations in the X column. Enter total binding for the first condition in column A and the corresponding nonspecific binding in column B. For the second condition, enter total binding in column C, and nonspecific in D. You can enter up to 26 pairs of data sets on one sheet. Enter all values as cpm.

2. Click the Analyze button. From the Data manipulation panel, choose Remove baseline.

3. The baseline values are in columns B, D, etc. You want to subtract, not divide, the baseline. Prism will calculate a table with A-B, C-D, E-F, etc.

4. Since nonspecific binding is almost always a linear function of ligand concentration, check "Assume the baseline is linear". In this case, Prism first uses linear regression to find the line that best fits the nonspecific values, and then subtracts the nonspecific values predicted from this line. This option works even if you have not measured nonspecific binding at every concentration. If you leave the option box unchecked, Prism subtracts each nonspecific value from the corresponding total value.

5. Check "Make new graph".

6. Click ok. The specific binding of the first data set will appear in column A of the results table. The specific binding of the second data set (originally columns C and D) will be in column B.

This method is only valid when a small fraction of the ligand binds to the receptor. If this assumption is true, then the free concentration of ligand equals the added concentration in both the tubes used to measure total binding and the tubes used to measure nonspecific binding. If the assumption is not valid, then the free concentration of ligand will differ in the two sets of tubes. In this case subtracting the two values makes no sense, and determining specific binding is difficult. Instead, you should fit total binding only as explained in Fitting  total binding data to determine Bmax and Kd.