This article describes knowledge-based systems for genetics called GENETICS-I, II, and III, and their possible extensions. GENETICS-I works on a simple genetic model where a phenotype is determined by a gene-pair, each gene having a value of 0 or 1 (diallelism). In GENETICS-II, which is a generalization of GENETICS-I, each gene can have a value of 0, 1,..., or gmax (multi-allelism; e.g. the ABO blood group). GENETICS-Ill is a further generalization of GENETICS-II in which a phenotype is determined by more than one pair of genes (multi-genes; e.g. major histocompatibility complex).
In each system, a knowledge base is established as a collection of production rules which are repeatedly applied on the database representing phenotypes and genotypes for a family tree to deduce new information. During the course of generalization, substantial changes in the database and knowledge-base structures have been made to deal with new types of problems as well as to increase the efficiency of computer time and memory space utilization. Possible extensions of these systems to include some common characteristics in expert systems are also discussed; included are a heuristic search of rules, user-system interactions, and reasoning under uncertain information.