Those who have a striking physical resemblance but aren’t related by blood often have comparable DNA and much more in common in terms of appearance, including body type, height, weight, and habits like smoking.

This is shown by the results of a fascinating research that analyzed 32 sets of twins from different parts of the globe, all of whom were recognized and shot by a Canadian artist. Cell Reports reports on the findings of a study headed by geneticist Manel Esteller of the University of Barcelona in Spain.

Esteller says, “For decades, the presence of people who appear identical without having any familial links has been portrayed as a verified truth,” but only in anecdotal terms and without scientific basis.

Today, we are able to identify and analyze these individuals because to the extensive use of the Internet and social networks for exchanging photographs.

Using photographs taken over the period of 20 years by Canadian artist Francois Brunelle, the researchers enlisted 32 sets of duplicates consisting of individuals who seemed to be twins but were really strangers to each other. After filling out a questionnaire about their biometrics and habits, all participants’ features were compared objectively using three different facial recognition algorithms to see how similar they were.

At least two programs were able to agree on the degree of resemblance between 25 out of 32 pairings of faces. In particular, all three methods agreed that 16 different pairings were equivalent. Saliva samples from the participants were analyzed to confirm that the 16 sets of twins share a high degree of genetic similarity (based on an evaluation of over 19,000 SNPs, the most common type of genetic variation among people), but that they differ in the mechanisms by which the genome is regulated (epigenetics) and the microbiome, both of which are heavily influenced by their respective environments. Additionally, several characteristics, both physical (such as weight and height) and behavioral (such as how sociable you tend to be)

Researchers contend that the work has potential implications in forensic medicine (to rebuild the researchers’ identikit based on their DNA) and in preventative screening, despite the limited number of duplicates analyzed. (by analyzing face features, we may guess which individuals may be carrying disease-causing mutations).