3D Face Shape Analysis for Biometry Forensic Investigation and Genetic Mapping

Abstract

Understanding the effect of genes on facial morphology has been essential in understanding craniofacial anomalies. The early diagnosis of possible syndromes through the face can be highly significant in early pediatric medical treatments. However, it gets more challenging when studying genes of correlation with normal range variations of facial traits. These correlations are of great interest for forensic scientists and anthropologists as retrieving face morphology via DNA analysis can be vital in identifying human remains, especially when reference samples are not available in the DNA database. The literature review shows that the current methods and tools are not ready yet to predict an entire face from DNA materials. Moreover, most studies on genetics and face morphology were conducted on the European population group. However, the Middle Eastern population is underrepresented in Genome public database. This bias could impair the accuracy of findings if direct associations were to be used across populations. Also, the potential for discovering novel genetic associations could be missed by not exploring other population face traits. Additionally, limited understanding of genetic differences between the Middle Eastern and the global populations will remain. This research was conducted on United Arab Emirates (UAE) nationals considered of Middle Eastern descent to bridge that gap. This thesis compares the genetic variations in the UAE population with previously reported Single Nucleotide polymorphisms (SNP)-face morphology associations. In addition, a Genome-Wide Association Studies (GWAS) study is conducted on the UAE cohort to identify genetic markers that are significantly associated with face measurements. Phenotypic data are retrieved semi automatically from the subjects' three-dimensional (3D) face scans. About half of the SNP-face selected associations were found to be in similar face areas compared to previous studies. SNPs' minor alleles in the UAE cohort were also mapped against five major population groups in the 1000 Genome Project. The results showed the highest concordance with the South Asian population group. Our results from the SNPs' study suggest that 3D face morphology complements the genetic association across different ethnic groups. A principal component analysis (PCA) was conducted on our cohort and the samples from the HapMap 3 project from the GWAS study. The PCA indicates the uniqueness of the UAE population and its admixed nature. The GWAS analysis reveals significant associations of 19 genetic loci with six face features, 14 of which are novel. Significant associations of five loci: ROR2, GALNTL6, RBFOX1, ERICH6-AS1, and RNU6-639P, with face morphology were previously reported in studies conducted on other population groups. Two loci, ROR2 and GALNTL6, are related to craniofacial malformations, demonstrating the importance of these genetic variations in face morphology. The results suggest the influence of historical migrations in the Middle Eastern region. This is the first study that investigates the association between 3D face traits and genetic variations in an Arab population to the best of our knowledge.

Date of Award	May 2022
Original language	American English