A study published in the Cell Report on August 23 found that people who are almost identical but not related are also surprisingly similar in genetic composition and tend to have the same lifestyle characteristics.

This result is not entirely surprising to Esteller, pointing out that genome-wide association studies have shown that some genetic mutations are related to facial features. Mario Falchi, who conducted twin studies at King's College, University of London, agreed, and he was not involved in the study. "The elegant experiments described in the paper ... show that individuals with similar appearances have more genetic variation than individuals selected randomly."

Esteler commented that the team also found that the similarities between the two exceeded their appearance and genes. He said that while the study was designed to examine the face, the team used “a very extensive questionnaire” that showed that physiological characteristics such as weight and height also had similarities in the secondary population, as did lifestyle characteristics such as smoking habits and education. This may suggest that the similarities “expand more traits related to personality,” he hypothesized, adding that such claims require “careful evaluation.” Falchi said it was reported that the association between facial features and physiological and behavioral phenotypes is likely due to multidirectional effects, where one gene affects multiple seemingly unrelated traits, while epistasis effects are the result of gene interactions. "It's interesting to see the link between genes and facial recognition algorithms; this suggests that using these two sets of data may be more powerful in the future." "This may open a Pandora's box for forensics , but it's exciting."

Esteller believes that this result may lead to the use of artificial intelligence to infer gene mutations based on face images. He speculated that scientists could eventually “distinguish from the nose or mouth that the person is a (medical-related) mutation carrier”, which could lead to early intervention to prevent the development of the disease and more targeted care.

The opposite result may also become reality, and the results may lead to better facial reconstruction DNA evidence, or as Esteller puts it, “how to draw a face from the genome.” A possible step in

in this direction, Esteller said, “there is a possibility for

in this direction, and Esteller said he and his team are now investigating the complete genome to discover other variations involving facial similarity. He also plans to study the role of proteome and transcriptome in facial changes. Using this multiomic approach, Esteller said, “We will be able to rebuild the perfect face from biomaterial .”