Lectures on Designing Screening Experiments

Abstract

Designing Screening Experiments (DSE) is a class of information - theoretical models for multiple - access channels (MAC). We discuss the combinatorial model of DSE called a disjunct channel model. This model is the most important for applications and closely connected with the superimposed code concept. We give a detailed survey of lower and upper bounds on the rate of superimposed codes. The best known constructions of superimposed codes are considered in paper. We also discuss the development of these codes (non-adaptive pooling designs) intended for the clone - library screening problem. We obtain lower and upper bounds on the rate of binary codes for the combinatorial model of DSE called an adder channel model. We also consider the concept of universal decoding for the probabilistic DSE model called a symmetric model of DSE.