Mapping genotypes into human face phenotypes

人脸是区分个体的关键特征。人脸的形状主要由遗传学决定，可能涉及发育、基因和调节因素。然而，人们对面部变异的基因组学知之甚少。因此，通过使用3D面部图像和基因组学来研究面部发育，我的长期目标是揭示人类面部的基因-表型联系，以及人类面部特征的遗传和表观遗传调节原理。*包括FaceBase(面部研究人员资源)在内的面部生物学社区已经生成并将生成大量数据集。这些包括人类/小鼠面部解剖区域及其发育前体的全基因组基因表达和基因调控图景，数千健康个体的3D面部图像和SNPs(单核苷酸多态)。卡尔加里大学的这个实验室是世界上十几个生成3D人脸数据的实验室之一。然而，整合和解释这些数据以进一步深入了解人脸正常范围变异的基因理解是具有挑战性的。*目前，3D面部建模关注的是整体面部特征，而面部GWAS(基因组范围关联研究)分析忽略了面部特征的发展背景。我们假设面部特征是由该特征的发育前体的基因调控网络(GRN)和信号网络(SNS)决定的。因此，同一性状的相同表型变化可能是由这些网络中的不同SNP引起的。因此，我这个研究计划的短期目标是：(I)开发计算工具来增强3D面部分析和网络建模，以及(Ii)开发一个新的GWAS框架(即DevNetGWAS)，该框架将面部特征的发育前体的GRN/SNS集成到GWAS分析中，以显著改善GWAS的统计关联，并推断该面部特征的遗传机制。*为了达到这些短期目标，我提出了以下五个项目：(I)在解剖学本体论的指导下，构建面部特征的发育前体特异性SNS和GRN，(Ii)开发用于以更详细的方式建模面部解剖区域(例如嘴唇)的新算法，(Iii)开发用于建模面部特征的发育前体特定网络的新算法，(Iv)开发面部特征的依赖于发育环境的GWA(即DevNetGWAS)框架，以及(V)将DevNetGwas应用于3D面部GWAS数据，以识别和实验验证面部特征(例如嘴唇)的遗传相关基因和网络模块。这些工具和DevNetGwas框架将为理解面部基因组学铺平道路，并推动人类特征Gwas分析领域的发展。

The human face is a critical feature that enables to distinguish individuals. The shape of the human face is mainly determined by genetics, probably involving development, genes, and regulatory elements. However, the genomics of facial variation is poorly understood. Therefore, by studying facial development using 3D facial images and genomics, my long-term goal is to uncover genotype-phenotype connections of the human face and the principles of genetic and epigenetic regulations in human facial traits. ***The facial biology community including FaceBase (A Resource for Facial Researchers) has generated and will generate extensive data sets. These include genome-wide gene expression and gene regulatory landscapes for human/mouse facial anatomic regions and their developmental precursors, 3D facial images and SNPs (single-nucleotide polymorphisms) for thousands of healthy individuals. The lab at the University of Calgary is one of a dozen labs in the world to generate 3D human facial data. However, it is challenging to integrate and interpret these data to further get insight into the genetic understanding of normal-range variations of the human face. ***At the moment, 3D facial modeling is focusing on global facial features, and the facial GWAS (genome-wide association study) analysis ignores the developmental context of facial traits. We hypothesize that a facial trait is determined by the gene regulatory networks (GRNs) and the signaling networks (SNs) of the developmental precursors of that trait. Therefore, a same phenotypic change of a trait can be caused by different SNPs in these networks. Therefore, my short-term goals of this research program are (i) to develop computational tools to enhance 3D facial analysis and network modeling, and (ii) to develop a new GWAS framework (i.e., DevNetGWAS) which integrates the GRNs/SNs of a facial trait's developmental precursors into the GWAS analysis for significantly improving the GWAS statistical associations, and inferring genetic mechanisms of that facial trait.***To reach these short-term goals, I have proposed five projects with the following objectives: (i) to construct developmental precursors-specific SNs and GRNs of a facial trait with the guidance of the anatomic ontology, (ii) to develop new algorithms for modeling facial anatomic regions such as the lip in a more detailed manner, (iii) to develop new algorithms for modeling the facial trait's developmental precursors-specific networks, (iv) developing a facial trait's developmental context-dependent GWAS (i.e., DevNetGWAS) framework, and (v) to apply the DevNetGWAS to the 3D facial GWAS data for identifying and experimentally validating genetically associated genes and network modules for a facial trait (eg, the lip). These tools and the DevNetGWAS framework will pave the way for the understanding of facial genomics and advance the field of human trait GWAS analysis.