Rapid serological profiling by enzyme-linked immunosorbent assay. II. Comparison of computational methods for...

An enzyme-linked immunosorbent assay (ELISA) was used to measure specific antibody activity from a single serum dilution in sera of chickens exposed to Newcastle disease virus (NDV). Observed endpoint titers were used to formulate regression equations, and then absorbance data obtained at a single serum dilution were converted directly to antibody titer by three methods: a correction factor method, a subtraction method, and a double-regression method. Each method was evaluated for three criteria: the overall stability of between-test antibody titer for control sera, the linearity of the relationship of the absorbance values at a single working dilution to the observed antibody titers, and the method's accuracy in predicting titers. Although a nearly linear relationship was obtained for all treatment methods examined, the double-regression method provided the best reduction of between-test titer variation and also best predicted titers.