The distinction between dominant and recessive alleles seems to be due to loss or gain of function in the proteins they code for. Alleles have been observed exhibiting incomplete and hierarchical dominance.
Another type is "circular dominance," which is modeled in some computer programs but it isn't clear whether this occurs in nature. In this case, the dominance of alleles works similarly to rock-paper-scissors.
Would circular dominance exist in nature? How would alleles determine circular dominance?