Optimal Two-Stage Algorithms for Group Testing Problems
SIAM Journal on Computing
Non-adaptive complex group testing with multiple positive sets
TAMC'11 Proceedings of the 8th annual conference on Theory and applications of models of computation
| Hi-index | 0.00 |
The combinatorial group testing problem is concerned with the design of experiments so as to minimize the number of tests needed to find the sets of items responsible for a particular phenotype (an observable property). The traditional group testing problem only considers the OR problem, i.e. the phenotype appears as long as one of the responsible items exists. In this paper, we introduce the phenotype testing problem which is a generalization of the well-studied combinatorial group testing problem. In practice, there are more than one phenotype and the responsible items and their mechanism (AND or OR) for each phenotype are unknown where an AND mechanism means that the phenotype only appears if all of the responsible items exist. This phenotype testing problem has an important application in biological research, known as phenotype knockout study. New algorithms for designing adaptive experiments for solving the phenotype testing problem with n items using O (logn ) tests in the worst cases are introduced. When the number of phenotypes is small, say at most 2, and the number of responsible items for each phenotype is at most 2, algorithms with near-optimal number of tests are presented.