形状形态测量分析：斑马鱼颅面形状量化的分析软件工具箱
2026年2月27日
本研究描述了一种利用免费软件（tpsDigs2、MorphoJ和PAST）量化颅面软骨形状的方案，用于测量斑马鱼幼体面部结构的变化。
我们的研究利用形态测量软件量化乙醇和bmp7a是否相互作用，导致斑马鱼的细微颅面缺损。线性度量会漏掉细微变化。通过这种方法，我们利用地标量化形状变化，同时控制尺寸差异。
首先，打开tpsDigs2软件。点击输入源码，然后选择文件，然后选择保存的TPS记事本文件。点击“选项”以可视化“图像工具选项。
将参考长度设为100微米，按下“设定刻度”。点击确定确认缩放参数，然后退出图片工具窗口。然后，点击瞄准符号，将第一个地标放置在梅克尔软骨之间的中线关节处。
第二个标志点位于梅克尔软骨与腭方骨之间的双侧关节。第三个标志点位于神经膜软骨之间的中线关节，第四个标志点位于腭方骨与神经膜软骨之间的双侧关节。然后，将第五个标志点放置在舌颌软骨的远端，确保每张图像的地标按同一顺序排列。
放置地标后，点击“文件”打开下拉菜单。选择保存数据，然后覆盖以保存更新后的文件。退出tpsDigs2软件。
启动MorphoJ软件。点击文件，选择创建新数据集，并为数据集分配名称。点击TPS，选择包含新添加数据点的记事本文件。
然后，创建数据集。在项目树中，点击数据集。选择前期，然后选择新普罗克鲁斯蒂斯合适。
选择按主轴对齐，点击“执行普罗克鲁斯斯特拟合”。在预备项中，选择生成协方差矩阵以获得前核坐标。在提示时执行该函数。
现在，选择创建或编辑线框，并链接图片上的点。点击链接点，接受或创建图像，然后根据需要编辑分类器。打开一个电子表格程序，同时保持MorphoJ开启。
输入分类器信息，如GENOTYPE、TREATMENT和GENOTYPE TREATMENT，然后将文件保存为CSV文件。在 MorphoJ 中，点击“文件”，选择导入分类器变量以导入 CSV 文件。返回项目树并点击数据集。
在“预言”中，选择编辑分类器以确认所有图片均已包含。在项目树中，选择CovMatrix，然后点击页面顶部的变体。选择主成分分析以计算主成分得分。
点击PC比分以显示生成的图表。右键点击图表，选择置信省略号以添加所需的分类器。选择颜色数据点，分配颜色，然后点击确定以应用更改。
在“初步”中，选择屏幕底部形状图的“设置选项”。选择线框图，修改目标形状、起始形状和数字的颜色。选择变异，然后选择前十字架方差分析。
将普罗克鲁斯的方差分析结果导出到之前创建的电子表格文件中。在MorphoJ中，选择项目树中的原始数据集。点击比较，然后选择典范变量分析。
选择分类变量GENOTYPE TREATMENT并执行该函数。要导出结果，点击“结果”标签，右键点击结果页面。选择导出为文件并保存结果。
在项目树中，选择正则变量分析，然后选择分数。点击文件，选择导出数据集，选择数据类型和基因类型处理。把CVA分数保存成TXT文件。
要准备文件以导入PAST软件，请在文本编辑器中打开已保存的CVA成绩。将左上角的术语id替换为Label，保存编辑后的文件。打开PAST软件，点击“文件”，然后选择编辑后的CVA成绩文件。
在导入窗口里。选择名称、行和列数据，Tab作为分隔符，然后点击导入。在“显示”标签下，选择列属性。
在“类型”旁的下拉菜单中，将“组”分配到包含分类器变量的第一列。突出显示主成分或典型变量数据列。选择多变量，然后选择检验，选择多变量正态性，分别运行主成分和典范变量数据的正态性检验。
点击上方左上角的灰色空格，选择整个数据集。选择多变量，然后选择“检验”，再选择多变量分析来进行分析。将MANOVA结果导出到之前创建的电子表格文件。
为每个基因型和处理组的每只幼虫拍摄了内脏颅骨图像。内脏颅骨软骨被标记为代表性图像，并在每张图像上放置地标，生成带地标的数据集。主成分1约占总形状变异的34%，主成分2占20%。
每个主成分相对于所有内脏颅骨的平均形状，在内脏颅骨形状上表现出特定变化。主成分分析图显示基因型组和处理组均值重叠，且无明显的聚类。乙醇处理的野生型幼虫由于样本量较小，平均值置信度大达95%。
典范变量1表示品红色线框中神经膜关节相较于平均黑色线框的细微缩短或延长。典型变异体2仅在内脏颅骨一侧，位于梅克尔氏-腭方格骨与腭方格-角膜关节之间的关节处出现内侧移位。乙醇处理的野生型幼虫表现出较大95%置信度椭圆，与其他组重叠。
多变量变异分析显示基因型和治疗对整体影响显著。样品的固定安装是最大的挑战。倾斜的幼虫可能导致测量不准确并影响结果。
该协议适应不同的解剖结构，分析三维数据，生成线性测量，并帮助确定实验样本量。未来研究将探讨其他基因型-乙醇敏感性，扩大样本量，并将这一多功能工具箱应用于各种解剖结构。
查看完整文字稿并访问数千部科学视频
Overview
This article presents a detailed morphometric protocol for analyzing subtle craniofacial shape changes in zebrafish larvae following ethanol exposure, with a focus on gene-ethanol interactions. The approach leverages landmark-based geometric morphometrics and multivariate statistical analyses to quantify and compare facial skeletal variation, addressing challenges in assessing fetal alcohol spectrum disorders (FASD) phenotypes.
Key Study Components
Area of Science
- Developmental biology
- Genetics
- Morphometrics
- Toxicology
Background
- FASD encompasses a range of physical, cognitive, and behavioral disabilities caused by prenatal ethanol exposure.
- Facial skeletal abnormalities are a hallmark of FASD, but their analysis is complicated by genetic variability and subtle shape changes.
- Zebrafish are an effective model due to genetic similarity to humans, external fertilization, conserved craniofacial structures, and transparent larvae.
- Traditional linear measurements may miss subtle or complex shape changes, necessitating advanced morphometric approaches.
Purpose of Study
- To quantify subtle craniofacial defects resulting from ethanol and bmp7a gene interactions in zebrafish.
- To develop and demonstrate a reproducible workflow for landmark-based morphometric analysis of craniofacial shape.
- To enable pairwise and multivariate comparisons of facial morphology across genotypes and treatments.
Methods Used
- Landmark placement on viscerocranial cartilage images using tpsDigs2 software.
- Procrustes alignment, principal component analysis (PCA), and canonical variate analysis (CVA) using MorphoJ.
- Multivariate normality and MANOVA tests using PAST software.
- Statistical visualization and export of results for further analysis.
Main Results
- Principal component 1 and 2 accounted for 34% and 20% of total shape variation, respectively.
- PCA plots showed overlapping means among genotype and treatment groups, with no distinct clustering.
- Canonical variate analysis revealed subtle, localized shape changes in the viscerocranium, such as shortening/lengthening of the ceratohyal joint and medial shifts at specific cartilage joints.
- MANOVA indicated a significant overall effect of genotype and treatment on craniofacial shape.
Conclusions
- The morphometric protocol enables detection of subtle craniofacial shape changes that linear measures may miss.
- Consistent sample mounting is critical to avoid measurement artifacts.
- This workflow is adaptable to other anatomical structures and can be extended to 3D data and larger sample sizes in future studies.
Frequently Asked Questions
Executive Industry Relevance
Quantitative morphometric analysis of craniofacial shape in zebrafish enables high-resolution detection of subtle phenotypic effects from gene-environment interactions, such as those relevant to fetal alcohol spectrum disorders. This workflow enhances predictive confidence in early-stage target validation and supports mechanistic de-risking for developmental toxicity studies. The approach is positioned to inform portfolio decisions where phenotypic screening and translational continuity are critical.
Strategic Applications in Biopharma R&D
Early Discovery & Target Validation
- Enables interrogation of gene-environment interactions affecting craniofacial development.
- Supports functional target validation by quantifying subtle morphological phenotypes.
- Facilitates mechanistic de-risking through multivariate statistical analysis of shape variation.
- Improves predictive confidence for developmental toxicity endpoints.
Screening & Assay Development
- Prepares validated zebrafish models for downstream phenotypic screening workflows.
- Standardizes landmark-based morphometric assays for reproducibility and scalability.
- Generates quantitative outputs suitable for comparative compound evaluation.
- Enables robust assay development for gene-environment interaction studies.
Translational & Preclinical Research
- Aligns zebrafish craniofacial phenotypes with disease-relevant endpoints in FASD research.
- Supports translational biomarker identification through quantitative shape metrics.
- Provides continuity from early discovery to preclinical validation of developmental toxicity.
- De-risks advancement decisions by clarifying subtle phenotypic effects.
Pipeline & Workflow Integration
This morphometric workflow integrates from early discovery through lead identification and preclinical research, supporting hypothesis testing and phenotypic screening in zebrafish models.
- Discovery Biology: Quantifies gene-environment effects on craniofacial development for hypothesis-driven studies.
- Screening: Delivers reproducible, quantitative shape data for assay readiness and compound evaluation.
- Analytics: Provides principal component, canonical variate, and MANOVA outputs for robust statistical comparison.
- Translational Research: Bridges zebrafish phenotypes to human-relevant developmental endpoints.
- Enterprise Reuse: Toolbox is adaptable to other anatomical structures and scalable for broader R&D applications.
Operational & Enterprise Impact
- Scientific Value: Increases predictive confidence and reduces mechanistic ambiguity in developmental toxicity studies.
- Operational Value: Standardizes morphometric analysis for reproducibility and cross-study comparability.
- Strategic Value: Informs go/no-go decisions and optimizes resource allocation in early-stage portfolios.
- Portfolio Impact: Enables risk-adjusted prioritization of gene-environment interaction targets.
Implementation Considerations
- Requires expertise in morphometric analysis and zebrafish developmental biology.
- Needs access to imaging, landmarking, and statistical analysis software (tpsDigs2, MorphoJ, PAST).
- Demands rigorous sample mounting and orientation for reproducible results.
- Adaptable to 3D data and other anatomical structures with protocol modifications.
- Sample size and mounting consistency are critical for statistical power and accuracy.
Biopharma FAQs
Chapters in this video
0:00
Introduction
0:24
Adding Landmarks to Zebrafish Larvae Viscerocranium Images Using tpsDigs2 Software
1:48
Using MorphoJ to Analyze Cranofacial Shape by Procrustes ANOVA
4:26
Canonical Variate Analysis of the Dataset and MANOVA in PAST Software
6:43
Results
8:30
Conclusion
