Introduction: Epigenetic Biometric Features

Genotype refers to a genetic constitution, or a group sharing it, and phenotype refers to the actual expression of a feature through the interaction of genotype, development, and environment. Genetic penetrance describes the heritability of factors, or the extent to which the features expressed are genetically determined. Those that are (such as blood group, gender, and DNA sequence) may be called genotypic features, and those that are not (such as iris patterns, compared above for genetically identical eyes) are phenotypic traits that are epigenetic (not determined genetically).

Persons who are genetically identical share all their genotypic features, such as gender, blood group, race, and DNA sequence. All biological characteristics of individuals can be placed somewhere along this "genotypic-phenotypic" continuum of genetic determination, with some features (e.g. gender; iris sequence) placed firmly at either endpoint. Other features such as facial appearance reveal both a genetic factor (hence identical twins "look identical") and an epigenetic factor (hence everyone's facial appearance changes over time).

Persons who share 50% of their genes (e.g. a parent and child; ordinary sibblings; fraternal twins; and double cousins) show a corresponding partial agreement in their genotypic features such as facial appearance at a given age, but no additional agreement in their epigenetic features.

The importance of these genetic aspects of biometric templates is that they directly influence the two basic error rates: False Match and False non-Match. Nearly one percent of persons have an identical twin, with whom they share all genotypic features such as their entire DNA sequence. This creates a minimum False Match rate of 1% (across a population) which we may call the biometric's genotypic error rate. Similarly, the tendency for some biometric features (such as facial appearance) to change over time creates a minimum rate of False Rejections, which we may call the biometric's phenotypic error rate. To maximize individuality, distinctiveness, and randomness, a biometric feature should be entirely epigenetic. To maximize stability over the life span, a biometric feature should not change with phenotypic development.

Comparisons Between Genetically Identical Iris Patterns

Although the striking visual similarity of identical twins reveals the genetic penetrance of facial appearance, a comparison of genetically identical irises reveals just the opposite for iris patterns: the iris sequence is an epigenetic phenotypic feature, not a genotypic feature. A convenient source of genetically identical irises are the right and left pair from any given person. Such pairs have the same genetic relationship as the four irises of two identical twins, or indeed in the probable future, the 2N irises of N human clones. Eye colour of course has high genetic penetrance, as does the overall statistical quality of the iris texture, but the textural details are uncorrelated and independent even in genetically identical pairs. This is shown in the Figure above, comparing 648 right/left iris pairs from 324 persons.

The mean Hamming Distance between genetically identical irises is 0.497 with standard deviation 0.031, which is statistically indistinguishable from comparisons between 9.1 million pairings of genetically unrelated irises (histogram provided here). This shows that the detailed phase structure extracted from irises by the phasor demodulation process is purely epigenetic, so performance is not limited (as it is for face recognition, DNA, and some other biometrics) by the birth rate of identical twins or by the existence of partial genetic relationships.


Back to Main Page.