A relational database model for describing DNA mutations is presented. The model was developed in conjunction with the human hprt database and was successful in representing over 1800 hprt mutations. Mutants showing aberrant mRNA splicing can be adequately described using the model, as well as mutants showing more than one mutation. The basic aspects of the relational model should be applicable to mutations in a variety of genes. A data entry program developed using Microsoft Access 2.0 is also described that implements the relational model. The data entry program ensures that relational integrity is maintained between the tables and automatically generates key fields as needed. The program also has the ability to convert between the various numbering schemes that are used to describe base pair location in the hprt gene. The program and source code are placed in the public domain so that other experimenters can adapt the program for use with other genes.