Martinez-Alonso J R, Zunzunegui-Pastor V, Marin-Rojas M C, Moreno-Gonzalez J, Millan-Santos I, Martinez-Sales V, Ortiz-Berrocal J
Comput Programs Biomed. 1976 Dec;6(4):249-62. doi: 10.1016/0010-468x(76)90041-6.
A method is devised for the automatic computation of data obtained from a series of determinations by radioimmunoassay or by competitive protein binding. The input data is introduced by means of an automatic paper tape punch attached to the sample counter. A series of programs is set up for a Hewlett Packard computer, model 2116B with 16K, which calculates bound and free fractions with control processes for varying possible errors. From experimental data, by linear least squares regression, a standard curve of best fit is derived and is considered in three different ways. The curve of best fit is selected by comparing the correlation coefficients obtained and by calculating the unknown sample concentrations. The program presents the results graphically together with a list of results which is for use within the Radiochemistry laboratory. Individual reports are also prepared for sending to those departments that requested the test.