On the statistical significance of primary structural features found in DNA-protein interaction sites

Probabilities of occurrence for a number of the symmetries and other sequence regularities found in DNA-protein interaction site sequences have been calculated for segments of random DNA sequence. Results show that many of the symmetrical and repetitive features seen in these interaction sites are likely to have occured by chance. Other features are so unlikely to have occurred by chance that they are probably involved in the DNA-protein interaction processes.