Computational anatomy and geometric shape analysis enables analysis of complex craniofacial phenotypes in zebrafish

Abstract

Computational anatomy (CA) approaches start by estimating a canonical template from a sample of images. This template is then used as a basis for statistical analysis to quantify structural differences among groups of interest. Due to the complexity of fish skulls, previous attempts to classify craniofacial phenotypes have relied on qualitative features or 2D landmarks. In this work we aim to identify and quantify differences in 3D craniofacial phenotypes in adult zebrafish mutants. We first estimate a synthetic normative zebrafish template using microCT scans from a sample pool of wildtype animals using the Advanced Normalization Tools (ANTs). To validate the accuracy of the template and our CA pipeline, we compared the otolith volumes derived from the CA approach to manually segmented volumes of the same set of zebrafish. Our overall CA based segmentation volumes are statistically similar to our manual segmentations and both results show that mutants have larger otoliths than their wildtype controls. We apply CA to zebrafish with somatic mutations in bmp1a, whose human ortholog when disrupted is associated with Osteogenesis Imperfecta. Our Generalized Procrustes Analysis separated the bmp1a and wildtype fish along the first two principal components which collectively explained 31.8% of the variation in the dataset. Compared to controls, the phenotypic differences in bmp1a fish are concentrated around the operculum and the orbit. Our CA approach seems to offer a potential pipeline for high throughput screening of complex fish craniofacial phenotypes, especially those of zebrafish which are an important model system for testing genome to phenome relationships in the study of development, evolution, and human diseases.

Examining craniofacial variation among crispant and mutant zebrafish models of human skeletal diseases

“…In two of the four model systems (plod2 and meox1) our pseudolandmark pipeline identified areas of variation in the skull. In contrast to manual landmarks, which are time intensive to collect and can introduce interobserver error into the dataset (Figure 1A), our results supports that our previously established pseudolandmark pipeline allows for more efficient data collection, higher consistency and repeatability, and denser morphometric analysis (Diamond et al, 2022).…”

“…In this study we tested a previously developed semiautomated pipeline (Diamond et al, 2022) to identify areas of craniofacial variation in four zebrafish models of skeletal disease. Our results suggest that regardless of the type of fish used in a study (somatic vs germline), there is a strong clutch effect in zebrafish.…”

“…With the exception of plod2 , in which the average of all landmarks was used as our manual gold standard, all genes were landmarked by a single author. Next, we transferred the set of 308 pseudolandmark points using the normative atlas of wildtype individuals that we previously published (Diamond et al, 2022) using the ALPACA module (Porto et al, 2021) of the SlicerMorph extension (Rolfe et al, 2021) in 3D Slicer (Fedorov et al, 2012). Pseudolandmark points were created using the PseudoLMGenerator module of the SlicerMorph package in 3D Slicer.…”

“…In this study, we build on a previously developed semiautomated pipeline (Diamond et al, 2022) to identify areas of craniofacial variation in four different zebrafish models of skeletal disease: plod2, meox1, sost and wnt16 . Zebrafish models of plod2 are associated with Bruck Syndrome, a rare form of Osteogenesis Imperfecta (Lv et al, 2018).…”

ANTsX neuroimaging-derived structural phenotypes of UK Biobank

“…Analogously, the Advanced Normalization Tools Ecosystem (ANTsX) is a collection of interrelated, open-source software libraries for biological and medical image processing and analysis [Tustison et al, 2021b] with developmental roots in high-performing medical image registration [Avants et al, 2011a; Avants et al, 2014] and built on the Insight Toolkit (ITK) [Yoo and Metaxas, 2005]. ANTsX-based IDPs have demonstrated utility in several studies spanning a variety of organ systems, species, and imaging modalities (e.g., [Diamond et al, 2022; Ding et al, 2022; Kini et al, 2021]). These IDPs include those which have been previously reported, such as global brain tissue volumes [Avants et al, 2011b] and more localized, FreeSurfer-analogous cortical thickness values [Tustison et al, 2014; Tustison et al, 2019; Tustison et al, 2021b] averaged over the Desikan-Killiany-Tourville (DKT) regions [Klein and Tourville, 2012].…”