Design and analysis of high-throughput screening experiments

Saturated row-column designs are studied to eliminate positional effects in primary highthroughput screening experiments. Compared to designs currently used in practice, all compounds in the designs considered are comparable within each microplate in spite of the existence of row and column effects. The designs considered in this paper also have the maximum number of compounds arranged in each microplate. Two statistical methods are used to choose leading compounds in the designs considered. These two methods take full advantages of effect sparsity in primary screening. Simulation studies are carried out to compare the two statistical methods with two ad hoc methods in selecting active compounds. A method that maintains balanced false positives and false negatives is recommended.