Have you ever heard of someone bumping into their doppelgänger and wondered how the forces of biology conspired to make two unrelated people look alike? In a recent study, a few familiar faces took advantage of a photographer’s keen eye and a multiomic “face-off” to find those factors that lead two strangers to look alike!
A previous study from the laboratory of Manel Esteller (Josep Carreras Leukaemia Research Institute, Spain) explored how epigenetic differences in monozygotic twins support differences in particular traits and disease penetrance. In their most recent study, the team approached this fascinating question from another angle – their multiomic face-off asked what genomic, epigenomic, and microbiomal factors determine facial similarities in unrelated human “look-alikes.”
So let’s hear from Joshi and colleagues about the multiomic face-off that described the factors influencing facial similarity:
- Three facial recognition algorithms – MatConvNet, Custom-Net, and Microsoft API – determine the objective measure of “likeness” for thirty-two human doubles recruited from the photographic work of François Brunelle
- This approach matches sixteen look-alike pairs, with likeness scores similar to those obtained from monozygotic twins
- Analysis of saliva DNA indicates that look-alikes share common genetic sequences
- Nine of sixteen look-alike pairs cluster together, with nearly 20,000 SNPs suggested as face trait variants that relate to body/face structures and associate with gene-expression changes
- The identified SNPs also impact other properties of the body and personality characteristics
- Analysis of saliva DNA methylation demonstrates that look-alikes possess diverse epigenomes, with DNA methylation patterns only matching one of the original sixteen look-alike pairs
- However, smaller epigenetic age differences exist among the nine “ultra-look-alike” compared with “non-ultra-look-alike” pairs
- Methylation quantitative trait loci (meQTLs), which influence methylation across genomic regions and can underlie direct SNP associations/gene-environment interactions, appear to underpin phenotypic commonalities in ultra-look-alike pairs
- Microbiome analysis via ribosomal RNA direct sequencing clustered one look-alike pair together, suggesting that they do not share a common microbiome; however, a relationship between the microbiome and fat content may contribute to look-alike phenotypes
Through this multiomic face-off, we catch a glimpse of how similar genotypes, but not epigenomes and microbiomes, play a significant determining role in human likeness. “These results will have future implications in forensic medicine — reconstructing the criminal’s face from DNA – and in genetic diagnosis – the photo of the patient’s face will already give you clues as to which genome he or she has,” senior author Manel Esteller says. “Through collaborative efforts, the ultimate challenge would be to predict the human face structure based on the individual’s multiomics landscape.”
Find a multiomic understanding of your very own doppelgänger by readying the original study at Cell Reports, August 2022.